Overview

Throughout the drug discovery and development process, target safety assessments help with understanding the role of the drug target in normal physiology and potential unintended adverse consequences and safety liabilities when modulating the target with a chemical compound or drug (Brennan, 2017).

To support target prioritisation, we have manually curated experimental data and insights from publications and other well-known sources of target safety and toxicity data, including the ToxCast, AOPWiki and PharmGKB (Open Targets downstream analysis of the toxicity datasets).

Safety data is available on the target profile page and can be accessed to provide a systematic view of potentially relevant target safety liabilities.

Computational pipelines and datasets

Target safety datasets are mapped to the correct Ensembl gene ID and ingested during our initial pipeline steps to enrich the target annotation object. The data is available for download as part of the target core annotation from our data download page.

Publications

Bowes J, Brown AJ, Hamon J, Jarolimek W, Sridhar A, Waldron G, Whitebread S. Reducing safety-related drug attrition: the use of in vitro pharmacological profiling. Nat Rev Drug Discov. 2012 Dec;11(12):909-22. doi: 10.1038/nrd3845. PMID: 23197038.

Brennan R.J. (2017) Target Safety Assessment: Strategies and Resources. In: Gautier JC. (eds) Drug Safety Evaluation. Methods in Molecular Biology, vol 1641. Humana Press, New York, NY. doi: 10.1007/978-1-4939-7172-5_12

Force T, Kolaja KL. Cardiotoxicity of kinase inhibitors: the prediction and translation of preclinical models to clinical outcomes. Nat Rev Drug Discov. 2011 Feb;10(2):111-26. doi: 10.1038/nrd3252. PMID: 21283106.

Ann M. Richard, Richard S. Judson, Keith A. Houck, Christopher M. Grulke, Patra Volarath, Inthirany Thillainadarajah, Chihae Yang, James Rathman, Matthew T. Martin, John F. Wambaugh, Thomas B. Knudsen, Jayaram Kancherla, Kamel Mansouri, Grace Patlewicz, Antony J. Williams, Stephen B. Little, Kevin M. Crofton, and Russell S. Thomas. ToxCast Chemical Landscape: Paving the Road to 21st Century Toxicology. Chemical Research in Toxicology 2016 29 (8), 1225-1251. doi: 10.1021/acs.chemrestox.6b00135. PMID: 27367298.

Lamore SD, Ahlberg E, Boyer S, Lamb ML, Hortigon-Vinagre MP, Rodriguez V, Smith GL, Sagemark J, Carlsson L, Bates SM, Choy AL, Stålring J, Scott CW, Peters MF. Deconvoluting Kinase Inhibitor Induced Cardiotoxicity. Toxicol Sci. 2017 Jul 1;158(1):213-226. doi: 10.1093/toxsci/kfx082. PMID: 28453775; PMCID: PMC5837613.

Lynch JJ 3rd, Van Vleet TR, Mittelstadt SW, Blomme EAG. Potential functional and pathological side effects related to off-target pharmacological activity. J Pharmacol Toxicol Methods. 2017 Sep;87:108-126. doi: 10.1016/j.vascn.2017.02.020. PMID: 28216264.

Urban L, Whitebread S, Hamon J et al. Screening for safety-relevant off-target affinities. In: Polypharmacology in Drug Discovery. Peters JU (Ed.). John Wiley and Sons, NJ, USA (2012). doi: 10.1002/9781118098141.ch2.

Welcome!

The Open Targets Platform is a comprehensive tool that supports systematic identification and prioritisation of potential therapeutic drug targets.

By integrating publicly available datasets including data generated by the Open Targets consortium, the Platform builds and scores target-disease associations to assist in drug target identification and prioritisation. It also integrates relevant annotation information about targets, diseases/phenotypes, drugs, variants, studies, and credible sets as well as their most relevant relationships.

The Platform is a freely available resource that is actively maintained with quarterly updates. Our data can be accessed through an intuitive web user interface, an API, and data downloads. Likewise, our pipeline and infrastructure codebases are open-source and can be used to create a self-hosted private instance of the Platform with custom data. For more information, please review our Licence documentation and, if you use our data and/or pipelines, please cite our latest publication.

Check out our blog to learn more about the Platform and the Open Targets research programme.

You can also join the Open Targets Community and follow us on:

• LinkedIn: Open Targets

• Bluesky: @opentargets.org

• Twitter: @opentargets

• YouTube: Open Targets

For additional help with the Open Targets Platform, or to report bugs, data issues, or submit a feature request, please post on the Open Targets Community, using the relevant categories and tags. If the request you would like to make has already been posted, please like the post to indicate you would like this to be prioritised.

Overview

To support target prioritisation, the Open Targets Platform includes tractability data that identifies key details, including if there is a binding site suitable for small molecule binding, an accessible epitope for antibody based therapy, relevant data for using Proteolysis Targeting Chimeras (PROTACs), or a compound in clinical trials with a modality other than small molecule or antibody.

The tractability data can assist in target prioritisation by identifying potential drug targets suitable for discovery pipelines and therapeutic modalities that are most likely to succeed. It also supports further investigation of targets for which there are no ligands or experimental structures or those targets outside a "druggable" target family but with strong genetic associations.

Our target tractability is based on a modified version of Approaches to target tractability assessment – a practical perspective and The PROTACtable genome and has workflows that generate tractability assessments for small molecule (SM), antibody (AB), Proteolysis Targeting Chimeras (PR), and other clinical (OC) modalities.

Assessments

The tractability assessments displayed on the Platform's target profile pages is the result of an open-source computational pipeline that performs in silico tractability assessments with small molecule, antibody, PROTAC, and other clinical modality workflows.

Data sources used in the pipeline include UniProt, HPA, PDBe, DrugEBIlity, ChEMBL, Pfam, InterPro, Complex Portal, DrugBank, Gene Ontology, and BioModels.

Assessments common to all modalities, ingested from ChEMBL, are:

• Approved Drug: the target has clinical precedence with Phase IV drugs

• Advanced Clinical: the target has clinical precedence with Phase II or III drugs

• Phase 1 Clinical: the target has clinical precedence with Phase I drugs.

We also include additional assessments specific to each modality.

Small molecule

• Structure with Ligand: Target has been co-crystallised with a small molecule (source: Protein Data Bank)

• High-Quality Ligand: Target with ligand(s) (PFI ≤ 7, SMART hits ≤ 2, scaffolds ≥ 2) (source: ChEMBL)

• High-Quality Pocket: Target has a DrugEBIlity score of ≥ 0.7 (source: DrugEBIlity)

• Med-Quality Pocket: Target has a DrugEBIlity score between 0 and 0.7 (source: DrugEBIlity)

• Druggable Family: Target is considered druggable as per Finan et al’s Druggable Genome pipeline.

Antibody

• UniProt loc high conf: High confidence that the subcellular location of the target is either plasma membrane, extracellular region/matrix, or secretion (source: Uniprot)

• GO CC high conf: High confidence that the subcellular location of the target is either plasma membrane, extracellular region/matrix, or secretion (source: Gene Ontology)

• UniProt loc med conf: Medium confidence that the subcellular location of the target is either plasma membrane, extracellular region/matrix, or secretion (source: Uniprot)

• UniProt SigP or TMHMM: Target has a predicted signal peptide or trans-membrane regions, and not destined to organelles (source: Uniprot SigP, TMHMM)

• GO CC med conf: Medium confidence that the subcellular location of the target is either plasma membrane, extracellular region/matrix, or secretion (source: Gene Ontology)

• Human Protein Atlas loc: High confidence that the target is located in the Plasma membrane (source: HPA)

PROTAC

• Literature: Target mentioned in a set of manually curated PROTAC-related publications (source: Europe PMC)

• UniProt Ubiquitination: Target tagged with the Uniprot keyword “Ubl conjugation [KW-0832]”, which indicates that the protein has a ubiquitination site, based on evidence from the literature (source: Uniprot)

• Database Ubiquitination: Target has reported ubiquitination sites in PhosphoSitePlus, mUbiSiDa (2013), or Kim et al. 2011

• Half-life Data: Target has available half-life data (source: Mathieson et al. 2018)

• Small Molecule Binder: Target has a reported small-molecule ligand in ChEMBL with a measured activity of at least 10 μM in a target-based assay (source: ChEMBL)

Computational pipeline and datasets

The data is available for download as part of the target core annotation from our data downloads page.

Alternatively, you can also download the input TSV file with the per-target assessments via FTP. To access this file, visit our FTP site and click on the release version (e.g. 21.04), followed by "input", followed by "annotation-files". You can then download the tractability_buckets TSV file. Descriptions of the columns found in the input file can be found on the pipeline README.md file.

Publications

Brown KK, Hann MM, Lakdawala AS, Santos R, Thomas PJ, Todd K. Approaches to target tractability assessment - a practical perspective. Medchemcomm. 2018 Feb 14;9(4):606-613. doi: 10.1039/c7md00633k. PMID: 30108951; PMCID: PMC6072525.

Schneider M, Radoux CJ, Hercules A, Ochoa D, Dunham I, Zalmas LP, Hessler G, Ruf S, Shanmugasundaram V, Hann MM, Thomas PJ, Queisser MA, Benowitz AB, Brown K, Leach AR. The PROTACtable genome. Nat Rev Drug Discov. 2021 Jul 20. doi: 10.1038/s41573-021-00245-x. PMID: 34285415.

Overview

The Molecular Interactions data aggregates and integrates interaction evidence reported in several resources to provide a systematic view on potentially relevant drug targets. Each of the integrated resources captures relationships of different nature including physical binary interactions, enzymatic reactions, or functional relationships. The information here available aims to capture not only the topology of the interaction network, but also the supporting experimental evidence reported on each of the databases.

In order to maximise coverage, the network contains all reported binary relationships between gene products (proteins and RNAs). Although the main focus are interactions between human molecules, the data also includes additional interactions between human gene products and molecules encoded in the genome of infectious pathogens (viruses and bacteria).

Data sources

IntAct

IntAct – http://www.ebi.ac.uk/intact – is a freely available, open source database for molecular interaction data. IntAct contains physical interactions derived from literature curation or direct user submissions.

Interactions are scored using the MI score. Benefiting from the PSI-MI controlled vocabulary, the Intact MI score provides a normalised (0 to 1) score that weights how recurrently an interaction has been reported, together with the confidence of the experimental techniques reported. Note that a high scoring interaction can be due to high-confidence evidence, but also a social bias on studying certain proteins. Generally speaking, scores > 0.4 correspond to medium to high confidence interactions, although some good-quality high-throughput interactions might still be scored below that threshold. More info on MI score can be found in the Intact documentation.

Interactions are grouped by interaction detection method and interaction type. As a consequence, the same pair of interactors might be split into multiple entries if individual proteins are reported to have different biological roles.

For IntAct, please note:

• The network only contains human and selected pathogen data from IntAct

• The majority of interactions are not directional and not signed. However, there are a proportion of interactions where the biological role of the participants can be stated and directionality specified (e.g. enzymatic reactions)

Reactome

Reactome – https://reactome.org/ – is an open source, open access, manually curated and peer-reviewed pathway database.

For Reactome, please note:

• Only human-human interactions are provided

• Interactions are directional and signed, with biological roles assigned to each participant if possible

• Protein interactions in Reactome are inferred from pathways and complexes based on Reactome internal method.

SIGNOR

SIGNOR, the SIGnaling Network Open Resource – https://signor.uniroma2.it/ – contains signaling information published in the scientific literature, which is manually curated and stored in a structured format.

For SIGNOR, please note:

• SIGNOR only contains human data

• Interactions are directional and signed, with biological roles assigned to each participant

• The network pulls information from the SIGNOR relations file

STRING

STRING – https://string-db.org – contains functionally interacting proteins. While most interactions in the other resources capture different types of physical interaction between molecules, functional interactions do not necessarily interact physically. Both direct (physical) and indirect (functional) associations are derived from computational predictions, from knowledge transfer between organisms, or from interactions aggregated from other (primary) databases. STRING interactions provide an overall combined_score, as well as each of the pieces of information that compose this score. More information on STRING scoring can be found on their documentation page.

Computational pipeline and datasets

The multipartite network displayed in the Open Targets Platform is the result of post-processing the information stored in a Neo4j graph database (graphDB). The graphDB does not provide STRING information but contains ComplexPortal information on stable protein complexes as an additional data source. The information of the graphDB is then exported together with STRINGdb and mapped to the Open Targets Platform targets (Ensembl Gene IDs).

The resulting dataset as well as all intermediate files can be found in the Open Targets Platform Data Access section or the Intact FTP.

Publications

When using this data please remember to acknowledge the sources:

Orchard S, Ammari M, Aranda B, Breuza L, Briganti L, Broackes-Carter F, Campbell NH, Chavali G, Chen C, del-Toro N, Duesbury M, Dumousseau M, Galeota E, Hinz U, Iannuccelli M, Jagannathan S, Jimenez R, Khadake J, Lagreid A, Licata L, Lovering RC, Meldal B, Melidoni AN, Milagros M, Peluso D, Perfetto L, Porras P, Raghunath A, Ricard-Blum S, Roechert B, Stutz A, Tognolli M, van Roey K, Cesareni G, Hermjakob H. The MIntAct project--IntAct as a common curation platform for 11 molecular interaction databases. Nucleic Acids Res. 2014 Jan;42(Database issue):D358-63. doi: 10.1093/nar/gkt1115. Epub 2013 Nov 13. PMID: 24234451; PMCID: PMC3965093.

Jassal B, Matthews L, Viteri G, Gong C, Lorente P, Fabregat A, Sidiropoulos K, Cook J, Gillespie M, Haw R, Loney F, May B, Milacic M, Rothfels K, Sevilla C, Shamovsky V, Shorser S, Varusai T, Weiser J, Wu G, Stein L, Hermjakob H, D'Eustachio P. The reactome pathway knowledgebase. Nucleic acids research. 2020 Jan;48(D1) D498-D503. doi: 10.1093/nar/gkz1031. PubMed PMID: 31691815. PubMed Central PMCID: PMC7145712.

Licata L, Lo Surdo P, Iannuccelli M, Palma A, Micarelli E, Perfetto L, Peluso D, Calderone A, Castagnoli L, Cesareni G. SIGNOR 2.0, the SIGnaling Network Open Resource 2.0: 2019 update. Nucleic Acids Res. 2020 Jan 8;48(D1):D504-D510. doi: 10.1093/nar/gkz949. PMID: 31665520; PMCID: PMC7145695.

Szklarczyk D, Kirsch R, Koutrouli M, Nastou K, Mehryary F, Hachilif R, Gable AL, Fang T, Doncheva NT, Pyysalo S, Bork P, Jensen LJ, von Mering C. The STRING database in 2023: protein-protein association networks and functional enrichment analyses for any sequenced genome of interest. Nucleic Acids Res. 2023 Jan 6;51(D1):D638-D646. doi: 10.1093/nar/gkac1000. PMID: 36370105; PMCID: PMC9825434.

Overview

Disease or phenotype annotation data sources

Identifying evidence implicating drug targets with diseases or phenotypes constitutes one of the pivotal challenges of the Open Targets Platform. Several steps are required to aggregate, integrate, validate, and score the collected information, in order to contextualise and weight the underlying evidence. The Platform provides a framework to allow the interactive or programmatic interrogation of target-disease evidence.

Data model

The Open Targets data model focuses on five main entities:

Entity annotation

These entities are annotated with a variety of public data sources, as well as information about the relationships between them. A more detailed description of the annotations is provided throughout the documentation. Some examples are:

Evidence generation and association scoring

Applications and data access

To start addressing therapeutic hypotheses users can find alternative ways to interface with the data:

Training materials

Online tutorial

Please note we are in the process of updating our training content following the most recent updates (25.03).

Webinar

Overview

In the Open Targets Platform, a study is a qualifying genome-wide association study for binary or quantitative traits.

Studies might capture different types of associations, such as curated associations (top hits) from literature or post-processing GWAS summary statistics.

For each study meeting the inclusion criteria, harmonisation, fine-mapping, colocalisation, and Locus-to-Gene analysis are performed, resulting in a list of significant credible sets and their most likely associated genes. The resulting 95% credible sets can be visualised in all study pages. Within the same study, credible sets might result from different fine-mapping pipelines but still be presented in the same study with their respective provenance and confidence.

Key study annotation

To provide a rich context to downstream interpretation of studies, and their associations, we capture a wide range annotations, whenever possible in a standardised way to enable interoperability.

For example:

• The measured trait/phenotype is standardised using the Experimental Factor Ontology (EFO), as the background trait when applied

• For molecular QTLs, the affected gene/protein is standardised to Ensembl gene identifiers

• For molecular QTLs, the biological context is standardised to relevant ontology (e.g. tissue — UBERON or cell ontology — CL)

• Depending on the granularity of the ingested study metadata, a rich cohort information is captured including sample sizes, ancestry composition and a list of named cohorts

• Indicated if association data of a study was ingested as summary statistics, together with the corresponding quality control measurements performed on the summary statistics

• Besides the Pubmed identifier, rich publication information is also captured including first author, publication year for easier data access

Study categories

Integrated studies can be split into two major groups that determine how their associations would be utilised in the target identification process.

1. Genome-wide association studies (GWAS)

A GWAS identifies associations between genetic variants and traits or phenotypes by analysing genome-wide variation data from a population. These studies cover binary or quantitative traits reported in the integrated sources.

Sources: GWAS Catalog, FinnGen

2. molQTL studies

A molQTL study identifies genetic variation that can be significantly associated with changes at the molecular level. As such, associations of these studies provide invaluable help to put GWAS derived loci into context via colocalisation. The type of molecular trait measured defines the type of QTL:

• eQTLs impact gene expression

• pQTLs impact protein abundance

• sQTLs have an effect on gene splicing

• tuQTLs impact transcript usage

• sceQTL impact gene expression at the single-cell level

Each molQTL study captures the unique combination of the following annotations:

• The publication authoring the study

• The gene product measured in the study

• The quantitative method (e.g. aptamer) used to measure the trait, when available

• The cell type or tissue where the trait is measured, when available

• The experimental conditions (e.g. interferon-stimulated macrophages)

Sources: eQTL Catalogue, UK Biobank Pharma Proteomics Project (UKB-PPP)

Study inclusion criteria

Studies must meet predefined criteria to ensure consistent representation, and increased reliability of associations for downstream analysis.

Each integrated study needs to meet the following criteria:

• GWAS studies need to have a valid disease or phenotype identifier (EFO)

• QTL studies must have an affected gene and tissue/cell-type valid identifier (biosample)

• Valid study type (e.g gwas or a QTL type from a defined set)

• Data licensed for commercial usage

• When summary statistics are available, further criteria must be met:

  • Mean beta within expected range

  • Genomic control (GC) lambda value within the expected range

  • The PZ value within the expected range

  • Additional curation from Open Targets team ensuring study meet standards

GWAS/molQTL credible sets

The fine-mapping results for all 95% credible sets in the study are displayed with their most relevant metadata, as well as the top Locus-to-Gene assignment.

Because the same study might be processed through different fine-mapping pipelines, it's possible to observe credible sets obtained with different methods even within the same study. The credible set exclusion criteria describes the rules applied to avoid duplicated credible sets resulting from separate fine-mapping pipelines.

Source: Open Targets

Overview

Core essential target genes are unlikely to tolerate inhibition and are susceptible to causing adverse events if modulated. This is crucial safety information for drug discovery scientists looking to develop inhibition strategies.

To support target prioritisation with an extra focus on target safety, the Open Targets Platform includes key information on target core essentiality in the context of 28 tissues assayed in the DepMap portal. In this project and its ancillary projects (e.g. Achilles), cell fitness is measured after the inhibition of individual genes across a number of cell lines. As a result, a gene is catalogued as core essential if the majority of cell lines die after inhibition/knockout. Although in cancer, this experiment represents a good proxy of whether loss of function is tolerated across a diverse set of tissues.

Essentiality assessment: The Chronos dependency score is based on data from a cell depletion assay. A lower Chronos score indicates a higher likelihood that the gene of interest is essential in a given cell line. A score of 0 indicates a gene is not essential; correspondingly -1 is comparable to the median of all pan-essential genes.

Together with a DepMap target annotation widget, essential target genes are annotated with a `Core essential gene` chip on the target page.

Data source

DepMap Portal

The goal of the Dependency Map (DepMap) portal is to empower the research community to make discoveries related to cancer vulnerabilities by providing open access to key cancer dependencies analytical and visualisation tools.

Publications

When using this data please remember to acknowledge the sources:

Pacini C, Dempster JM, Boyle I, Gonçalves E, Najgebauer H, Karakoc E, van der Meer D, Barthorpe A, Lightfoot H, Jaaks P, McFarland JM, Garnett MJ, Tsherniak A, Iorio F. Integrated cross-study datasets of genetic dependencies in cancer. Nat Commun. 2021 Mar 12;12(1):1661. doi: 10.1038/s41467-021-21898-7. PMID: 33712601; PMCID: PMC7955067.

See all relevant DepMap publications.

Overview

To support further assessments about the suitability of targets for a discovery pipeline, the Open Targets Platform integrates information from various sources on chemical probes and Target Enabling Packages (TEPs).

Chemical probes

A chemical probe is a small molecule that can act as chemical modulator of a system, such as a cell or an organism, by reversibly binding to a biological target. Chemical probes are expected to have a minimum standard of high affinity, binding selectivity and efficacy. Chemical probes do not need to meet the same requirements in terms of pharmacokinetics, pharmacodynamics, and bioavailability as drugs.

Target Enabling Packages

Target Enabling Packages (TEPs) are a critical mass of reagents and knowledge allowing for the rapid biochemical and chemical exploration of a given target.

As noted by the Structural Genomics Consortium, each TEP contains:

• Protein production methods

• Biochemical/biophysical assays for activity, affinity

• Structures of the protein, potentially including wild type and disease mutant proteins; full-length or domains; protein-ligand complexes; structures of close homologues

• Initial chemical matter from a fragment or small molecule screen

Additional components of TEPs may be developed on a case-by-case basis, based upon reasonable scientific need, in collaboration with TEP target nominators:

• An antibody or nanobody

• Cell-based assay

• CRISPR knockout

Data sources

Probes & Drugs Portal

The Probes & Drugs Portal – https://www.probes-drugs.org/home/ – is a public resource joining together focused libraries of bioactive compounds (probes, drugs, specific inhibitor sets etc.) with commercially available screening libraries. The purpose of the Portal is to reflect the current state of bioactive compound space and to enable its exploration from different points of view. It is intended to serve as a central hub in chemical biology research by providing a unique integration of the most relevant probes sources such as the Chemical Probes Portal, Open Science Probes, and ProbeMiner.

Structural Genomics Consortium

The Structural Genomics Consortium (SGC) – https://www.thesgc.org/ – is a public-private partnership focused on accelerating drug discovery using open science. The SGC’s core research operations are funded by pharmaceutical companies, governments, and charities, all of which act as research partners and participate in the governance of the SGC.

Computational pipelines and datasets

Chemical probes datasets are downloaded from each data source listed above, mapped to the correct Ensembl gene ID, and ingested during our initial pipeline steps. The data is available for download as part of the target core annotation from our data download page.

Overview

Baseline RNA and protein expression data helps us to ascertain whether the target is expressed in all tissues or selectively in one or few tissues or cell types. The availability of the target molecule in the location of interest is critical at different stages of the drug development process.

We combine baseline expression information from three sources:

• Expression Atlas: RNA expression meta-analysis from RNA- sequencing experiments

• Human Protein Atlas: immunohistochemistry-based proteomics data for normal tissues

• Genotype-Tissue Expression (GTEx) Program: RNA baseline expression variation data per tissue

Data sources

Expression Atlas

Expression Atlas – https://www.ebi.ac.uk/gxa/home – is a manually-curated, freely available data analysis and visualisation resource that provides information about gene and protein expression in more than 3,000 experiments.

Human Protein Atlas

The Human Protein Atlas – https://www.proteinatlas.org – is an international collaboration that develops various open access knowledge resources that map all the human proteins in cells, tissues and organs using an integration of various 'omics technologies. The Human Protein Atlas consists of six separate parts, each focusing on a particular aspect of the genome-wide analysis of the human proteins. At the moment, we only use the Tissue Atlas, showing the distribution of the proteins across all major tissues and organs in the human body.

Genotype-Tissue Expression (GTEx) Portal

The Genotype-Tissue Expression (GTEx) Portal – https://gtexportal.org/home/ – is an open access data resource that provides data on gene expression, QTLs, and histology images. The Portal is part of the GTEx consortium's ongoing effort to build a comprehensive public resource to study tissue-specific gene expression and regulation and its relationship to genetic variation.

Computational pipelines and datasets

RNA expression meta-analysis

Open Targets in collaboration with Expression Atlas has performed a meta-analysis of over 18,000 samples from 50 different tissues and more than 30 cell types from the following experiments:

• RNA-seq of 53 human tissue samples from GTEx (E-MTAB-5214)

• RNA-seq of 16 human tissues from the Illumina Body Map project (E-MTAB-513)

• RNA-seq of 13 human tissue from the ENCODE project (Snyder Lab) (E-MTAB-4344)

• RNA-seq of 6 human tissues from Kaessmann Lab (E-MTAB-3716)

• mRNA-seq of 32 human tissues from Human Protein Atlas (E-MTAB-2836)

• mRNA-seq of rare types of cells of different haemopoietic lineages from healthy individuals in the BLUEPRINT project (E-MTAB-3819)

• RNA-seq of common types of cells of different haemopoietic lineages from healthy individuals in the BLUEPRINT project (E-MTAB-3827)

• mRNA-seq of plasma cells of tonsil from healthy individuals from the BLUEPRINT project (E-MTAB-4754)

The tissue- and cell-based samples in these experiments are processed separately to avoid batch effects during normalisation. The samples of each group are then processed together to generate an expression table of normalised Transcripts Per Million (TPMs) units for every gene in each tissue or cell type according to the following steps:

• Technical replicates are aggregated

• Genes that are expressed below a pre-defined threshold (i.e. minimum of 10 raw reads in at least 15 samples) are filtered out

• Samples are then normalised in a two-step process according to Risso et al. 2014 using the RUV (Remove Unwanted Variation) method:

  • The Coefficient of Variation (CV) was estimated for each gene across all the samples and used to select the least variable genes.

  • The least variable 1,000 genes were used as negative controls; that is, assumed not to be differentially expressed, to train RUVg to remove unwanted variation

• Tissues are mapped to anatomical ontology terms using the Uber-anatomy ontology. Tissues that can't be mapped are then discarded

• Samples from the same tissue across different experiments are averaged by median and then merged in the final matrix

• Expression tables of the tissue- and cell-based experiments are combined

We analyse this expression file further to compute two values for each gene:

• Binned value of expression: The normalised expression values are divided into 10 bins of the same width. Note that this is not the same as the deciles, which all contain the same number of items in them

• Tissue specificity: Z-scores are calculated for each gene and each tissue and then they are binned based on quantiles of a perfect normal distribution. We compute the tissue specificity of a target as the number of standard deviations from the mean of the log RNA expression of the target across the available tissues. A target is considered to be tissue specific if the z-score is greater than 0.674 (or the 75th percentile of a perfect normal distribution). We remove data for under-expressed targets before the z-score calculation. This allows us to extract the tissues for which a gene is specific, defined as the expression value being above the 75th z-score percentile — in practice, anything in bin 2 or above.

Normalised files with binned value of expression and tissue specificity values for each target are available from our data download page.

Publications

GTEx Consortium. The Genotype-Tissue Expression (GTEx) project. Nat Genet. 2013 Jun;45(6):580-5. doi: 10.1038/ng.2653. PMID: 23715323; PMCID: PMC4010069.

Papatheodorou I, Moreno P, Manning J, Fuentes AM, George N, Fexova S, Fonseca NA, Füllgrabe A, Green M, Huang N, Huerta L, Iqbal H, Jianu M, Mohammed S, Zhao L, Jarnuczak AF, Jupp S, Marioni J, Meyer K, Petryszak R, Prada Medina CA, Talavera-López C, Teichmann S, Vizcaino JA, Brazma A. Expression Atlas update: from tissues to single cells. Nucleic Acids Res. 2020 Jan 8;48(D1):D77-D83. doi: 10.1093/nar/gkz947. PMID: 31665515; PMCID: PMC7145605.

Uhlén M, Fagerberg L, Hallström BM, Lindskog C, Oksvold P, Mardinoglu A, Sivertsson Å, Kampf C, Sjöstedt E, Asplund A, Olsson I, Edlund K, Lundberg E, Navani S, Szigyarto CA, Odeberg J, Djureinovic D, Takanen JO, Hober S, Alm T, Edqvist PH, Berling H, Tegel H, Mulder J, Rockberg J, Nilsson P, Schwenk JM, Hamsten M, von Feilitzen K, Forsberg M, Persson L, Johansson F, Zwahlen M, von Heijne G, Nielsen J, Pontén F. Proteomics. Tissue-based map of the human proteome. Science. 2015 Jan 23;347(6220):1260419. doi: 10.1126/science.1260419. PMID: 25613900.

Overview

A target in the Platform is understood as any naturally-occurring molecule that can be targeted by a medicinal product. EMBL-EBI's Ensembl database is used as source for human targets in the Platform, with the Ensembl gene ID as the primary identifier.

Criteria for target inclusion:

• Genes from all biotypes encoded in canonical chromosomes

• Genes in alternative assemblies encoding for a reviewed protein product.

This definition accounts for some of the complexities of human targets; targets are not only protein coding genes, RNAs or pseudogenes are also considered. However, the current definition has some potential drawbacks. Some drug targets are the result of interactions between genes (e.g. gene fusion) or proteins (e.g. protein complexes). The current target entity does not yet consider these cases, which will be addressed in the future.

Target annotation data sources

Overview

Pharmacogenetics is the study of how genetic variation may change your response to a specific drug. Through the integration of pharmacogenetics data into the Platform, our objective is to apply clinical annotations available in the Pharmacogenetics Knowledgebase (PharmGKB) to aid target prioritisation for drug discovery.

We also enhance these annotations by adding detailed annotations on variant consequence prediction, drug response categories, and specific drug information and whether the gene is a direct target of the drug. This process involves applying advanced phenotype extraction techniques to offer a refined representation of phenotypes, thereby providing a more precise understanding of the genetic determinants influencing treatment outcomes.

Data sources

We have also included annotations for star (*) alleles, a nomenclature used in the pharmacogenetics field for representing key functional variants involved in drug responses.

Publications

When using this data please remember to acknowledge the sources:

Whirl-Carrillo M, Huddart R, Gong L, Sangkuhl K, Thorn CF, Whaley R, Klein TE. An Evidence-Based Framework for Evaluating Pharmacogenomics Knowledge for Personalized Medicine. Clin Pharmacol Ther. 2021 Sep;110(3):563-572. doi: 10.1002/cpt.2350. Epub 2021 Jul 22. PMID: 34216021; PMCID: PMC8457105.

Whirl-Carrillo M, McDonagh EM, Hebert JM, Gong L, Sangkuhl K, Thorn CF, Altman RB, Klein TE. Pharmacogenomics knowledge for personalized medicine. Clin Pharmacol Ther. 2012 Oct;92(4):414-7. doi: 10.1038/clpt.2012.96. PMID: 22992668; PMCID: PMC3660037.

Overview

To further refine how we define a drug in the Platform, we subset ChEMBL's drugbase to capture molecules that meet the following criteria:

• Drugs for which the indication is known

• Drugs for which the target they modulate is identified

• Drugs that are listed in DrugBank

• Drugs that are acknowledged as chemical probes

A drug in the Platform might belong to different modalities, including small molecules, antibodies, or oligonucleotides among others. However, some biologic therapies such as vaccines, blood products, or cell therapies are not represented in our drug set. Moreover, the molecule-centric definition implies multi-ingredient drugs won't be represented and only their individual active moieties might be available on the site.

In the ChEMBL representation of drugs, a clear distinction is made between parent bioactive molecules and their corresponding child molecules. Parent molecules encompass the original, unmodified form of the active ingredient, while child molecules refer to modified versions, such as salts. In the Platform, both parent and child molecules are included, ensuring comprehensive coverage of the drug landscape.

When it comes to data propagation, parent molecules retain their own distinct information, as well as aggregate the specific details from all their child molecules. On the other hand, child molecules solely capture their own individual information, without incorporating any data from other molecules. This consistent approach extends to various aspects, including indications, mechanism of action, and drug warnings. By adopting this systematic framework, our Platform facilitates accurate representation and analysis of the diverse molecular entities and their associated properties.

Drug annotation data sources

Overview

Before approval, new therapeutic drug treatments are extensively tested in clinical trials. However, some of the side effects are only identified when prescribed to larger cohorts of patients, with one or more medical conditions, for a sustained period of time or in combination with other treatments.

For this reason, regulatory agencies—for example, the Food & Drug Administration or the European Medicines Agency—provide pharmacovigilance programs to monitor and survey Adverse Drug Reactions (ADRs).

Data sources

FDA Adverse Event Reporting System (FAERS)

Computational pipelines and datasets

First we apply a set of filters to the reports as described below:

• Only reports submitted by health professionals (primarysource.qualification in (1,2,3))

• Exclude reports that resulted in death (no entries with seriousnessdeath=1)

• Only drugs that were considered by the reporter to be the cause of the event (drugcharacterization=1)

Next, we sought to map the drugs in the FAERS reports to the drugs in the Open Targets Platform (ChEMBL IDs). Any of the above listed fields were used when exact matches were available:

Due to the nature of the surveillance reports, it's relatively common for the indication for which a drug was prescribed to appear in the list of significant ADRs. Given the current structure of the data provided in a FAERS report, we cannot distinguish whether it's a problem with the dosage the drug was prescribed or an excessive phenotypic characterisation of the patient in the report.

Publications

Overview

The Clinical Signs and Symptoms data in the Platform aims to capture other diseases or phenotypes that occur as a consequence, or in conjunction with a primary disease.

Disease–phenotype relationships are not only useful to better characterise the disease phenotypic space, but also to serve as proxies for additional causal evidence that can help prioritise new or existing targets.

Data sources

Monarch Merged Disease Ontology (MONDO)

Human Phenotype Ontology (HPO)

Computational pipelines and datasets

Publications

Overview

Population Allele Frequencies

Variant effect

Variants are annotated with an integrated view of variant effects from multiple methods. Based on all predictions or annotations, we normalise the variant's likely deleteriousness to a common scale.

To make the predicted variant effects comparable across different methods, raw predictions from each methods were normalised to a unified scale ranging from likely benign to uncertain to likely deleterious.

Transcript consequences

Every variant is annotated with the predicted consequence for all canonical transcripts in a +/-500Kb window, allowing us to understand the likely effects in the neighbouring coding or non-coding genes. For all variant-transcript pairs in the region, this information includes:

• Distance from transcription start site (TSS)

• Distance to footprint

• Predicted functional consequence based on Ensembl VEP

• Amino-acid consequence relative to the UniProt reference protein

Variant-to-phenotype

The list of variant sources includes:

The 'Clinical Precedence' widget in the Open Targets Platform provides users with an overview of the clinical development stage of each drug, as defined by the EMBL-EBI ChEMBL database. This widget is designed to aid in the prioritisation of drug targets by highlighting the extent of clinical development for each drug.

The clinical development stages are categorised as follows:

In the Platform, this information offers valuable insights into the clinical development of potential drug targets. By showing the current phase of clinical development, the 'Clinical Precedence' widget can inform decisions about target prioritisation and help identify drugs that may be suitable for repurposing or further development.

Overview

Pharmacogenetics is the study of how genetic variation may change your response to a specific drug. Through the integration of pharmacogenetics data into the Platform, our objective is to apply clinical annotations available in the Pharmacogenetics Knowledgebase (PharmGKB) to aid target prioritisation for drug discovery.

We also enhance these annotations by adding detailed annotations on variant consequence prediction, drug response categories, and specific drug information and whether the gene is a direct target of the drug. This process involves applying advanced phenotype extraction techniques to offer a refined representation of phenotypes, thereby providing a more precise understanding of the genetic determinants influencing treatment outcomes.

Data sources

We have also included annotations for star (*) alleles, a nomenclature used in the pharmacogenetics field for representing key functional variants involved in drug responses.

Publications

When using this data please remember to acknowledge the sources:

Whirl-Carrillo M, Huddart R, Gong L, Sangkuhl K, Thorn CF, Whaley R, Klein TE. An Evidence-Based Framework for Evaluating Pharmacogenomics Knowledge for Personalized Medicine. Clin Pharmacol Ther. 2021 Sep;110(3):563-572. doi: 10.1002/cpt.2350. Epub 2021 Jul 22. PMID: 34216021; PMCID: PMC8457105.

Whirl-Carrillo M, McDonagh EM, Hebert JM, Gong L, Sangkuhl K, Thorn CF, Altman RB, Klein TE. Pharmacogenomics knowledge for personalized medicine. Clin Pharmacol Ther. 2012 Oct;92(4):414-7. doi: 10.1038/clpt.2012.96. PMID: 22992668; PMCID: PMC3660037.

The Platform is built upon a significant effort to analyse genome-wide associations studies and interpret them in the context of functional genomics studies. This effort to inform target identification and prioritisation datasets leverages gentropy a highly-scalable python framework for post-GWAS analysis.

A more detailed view on the data and methodology is available in:

• Data sources used to ingest variant annotation, GWAS and functional genomics information.

• Fine-mapping pipelines to identify likely causal variants in GWAS-significant signals in different sources

• Colocalisation performed on overlapping GWAS and molQTL credible sets

• Locus-to-Gene predictions to assing likely causal genes to identified credible sets

The Platform ingests GWAS, functional genomics and additional datasets to aid the interpretation of the observed signals. These include:

NHGRI-EBI GWAS Catalog

🌍 Website — 📖 Gentropy

The NHGRI-EBI GWAS Catalog is a data source that provides detailed, structured genome-wide association study data in standardised format of summary statistics and curated associations (top-hits) to EFO traits. We rely on the GWAS Catalog for a rich source of genetic associations, utilising the data for analysis and interpretation.

The GWAS Catalog information feeds the Open Targets Platform by providing study metadata and GWAS associations from two sources:

GWAS Catalog literature-curated associations (GCCA)

The Platform ingests the GWAS Catalog curated associations (top-hits) from the table “All associations - with added ontology annotations, GWAS Catalog study accession numbers and genotyping technology” (see GWAS Catalog download page). The GWAS Catalog curation process may result in multiple GWAS being grouped under one study accession, in which case one study accession will need to be split into multiple studies in the Platform. An example of such a study is: GCST001762, where the reported trait is 'Obesity-related traits', but the underlying association data informs which top-hit comes from which specific trait, for example 'BMI z-score change'. In this case, we create as many new studies as there are association-level trait annotations. The new study identifiers are generated from the original study accession plus a suffix. The disease annotations of the newly created studies are inherited from the association data. A similar action is required when results from different ancestries are pooled under one study accession.

The harmonisation process includes a number of steps that flag any associations with quality concerns. When harmonising GCCA, the following cases were flagged:

• Variant interaction associations

• Variant location is not available in an association source

• Variant location data is inconsistent (chromosome and position values don’t match)

• Lead variant is duplicated for the same study

• The provided variant location doesn’t match any GnomAD variant

• The risk allele couldn't be mapped to the GnomAD reference

• The lead variant had palindromic alleles

• The lead variant p-value was above the genome-wide significant threshold (p-value>1e-8).

Curated associations were converted into the Gentropy StudyLocus format and uniformly flagged with Study locus from curated top hit quality control flag. Corresponding StudyIndex are generated to captures the study metadata.

GWAS Catalog Summary Statistics (GCSS)

The Platform ingests GWAS Catalog studies from full summary statistics when they have been harmonised following the protocol. Each harmonised study is converted into a Gentropy SummaryStatistics format as described in the Gentropy documentation including additional quality controls:

1. filtering out SNPs with unavailable beta, standard error, or p-value

2. filtering out SNPs with zero or Inf values for beta and standard error

3. filtering out SNPs with negative p-values or standard error

4. filtering out SNPs with p-values equal to 1

In some cases, this results in empty SummaryStatisticsdatasets and these studies are excluded from further analysis.

GCSS studies manual curation

The Open Targets team performs additional manual curation for the GCSS studies from “All studies - With study accession numbers, ontology annotations, genotyping technology, cohort identifiers and full summary statistics availability”. The following information is extracted from the corresponding publication:

1. Study type — The GCSS studies identified as pQTL or microbiome GWAS are flagged and excluded from further analysis.

2. Analysis type — studies are flag if performed with any of the following analysis: multivariate analysis, ExWAS, non-additive model, metabolite, GxG, GxE, case-case study. All analysis flags but “metabolite” are excluded from SuSiE fine-mapping and fine-mapped using PICS.

A StudyIndexdataset was created as part of this pipeline containing all the available metadata for the included GWAS Catalog studies. If available, the ancestries from GWAS Catalog were mapped to gnomAD ancestry suffixes using the dictionary here. The ancestry wasn't assigned if the ancestry was unavailable or wasn't presented in the dictionary.

GCSS quality control

GWAS summary statistics quality control (QC) is performed for all GWAS Catalog studies with available summary statistics, following the methods described in Winkler et al. (2014):

1. The P-Z test. This check estimates the mean and standard deviation of the difference between the log p-values reported in the study and the reported betas and standard errors. If at least one value for the study was greater than 0.05, the study fails QC and is flagged with the label The PZ QC check values are not within the expected range.

2. The mean beta check. This check estimates the mean value of the beta across all SNPs in the study. If the absolute mean value is more than 0.05, the study fails QC and is flagged with the label The mean beta QC check value is not within the expected range.

3. The genomic control (GC) lambda check. This check estimates the additive GC lambda of the study (see Tsepilov et al. (2013)). If the GC lambda value is outside the [0.7,2.5] range the study fails QC and is flagged with the label The GC lambda value is not within the expected range.

4. Number of variants. All summary statistics with fewer than 2,000,000 variants do not fail QC but are flagged with the label The number of SNPs in the study is below the expected threshold. These studies are not eligible for SuSiE fine-mapping.

FinnGen

🌍 Website — 📖 Gentropy

FinnGen is an academia-industry partnership that aims to produce genome variant data for 500,000 Finns. The genomic data is then combined with phenotype data collected by national health registries, including extensive longitudinal registry data available on all Finns. More details on the protocols are available in the FinnGen documentation. Because the Finnish population has been genetically isolated, fine-mapping was performed with a suitable reference panel of LD. The Platform includes the fine-mapping results from the FinnGen team using SuSiE and FINEMAP, based on a reference panel of whole genome sequencing data from Finns.

The Platform includes the 95% credible sets as described in the documentation after converting them Gentropy StudyIndex and StudyLocus objects. Credible sets with a lead SNP p-value>=1e-5 are marked with a Subsignificant p-value flag.

eQTL Catalogue

🌍 Website — 📖 Gentropy

The eQTL Catalogue aims to provide unified gene expression, protein-level and splicing QTLs from publicly available human public studies. Using a standardised pipeline (QTLmap), the eQTL catalogue is integrated in the Platform as a rich source of molecular QTLs (molQTLs). The pipeline uses SuSiE as the method of fine-mapping and results in 95% CSs.

All credible sets and study information are reformatted to Gentropy's StudyIndex and StudyLocus datasets. Credible sets with a lead SNP p-value>=1e-3 are marked with a Subsignificant p-value flag.

The Pharma Proteomics UK Biobank Project (PPP-UKBB)

📖 Gentropy

The Pharma Proteomics Project is a pre-competitive biopharmaceutical consortium characterising the plasma proteomic profiles of 54,219 participants in the UK Biobank.

The Platform includes GWAS full summary statistics for 2,954 proteins of European ancestry from the Synapse platform. Data is converted to SummaryStatistics and StudyIndex format applying the following modifications:

• SNPs with MAF<1e-4 and INFO<0.8 are filtered out

• Align the order of effective and reference alleles with the gnomAD annotation: if the alleles are reversed, the sign of the effect size is changed; if the allele combination do not match the reference, the SNP is filtered out

• Flag QTLs as cis or trans based on a 5Mb window from the affected gene.

gnomAD

🌍Website — 📖 Gentropy

gnomAD (Genome Aggregation Database) is a comprehensive resource that provides aggregated genomic data from large-scale sequencing projects. It encompasses variants from diverse populations and is widely used for variant annotation and population genetics studies.

Variant annotation

GnomAD (v4) variant annotation is used to provide extra annotation for variants included in the Platform and assist some of our harmonisation pipelines. Among the most relevant annotations included are: population allelic frequencies, in silico predictors and cross-references. All variants are available in GRCh38 coordinates.

LD matrices

gnomAD v2.1.1 LD matrices are used to create a multi-ancestry LD reference for the PICS fine-mapping method. All variants are liftovered to build GRCh38.

Pan-UKBB LD matrices

🌍 Website

The Platform uses Pan-UK Biobank project LD matrices for three ancestries: Non-Finnish European (NFE), Central/South Asian (CSA) and African (AFR). See the descriptive summary from Pan-UK Biobank. The LD was computed for each chromosome in 10 Mb radius. Only SNPs with INFO > 0.8, MAC > 20 in each population are used to calculate LD. Matrices are stored in hail BlockMatrix format. We use these LD matrices for SuSiE fine-mapping.

EMBL-EBI Ensembl

🌍 Website — 📖 Gentropy

The Platform uses Ensembl as a source of gene, transcript and variant annotations. Affected genes in QTL studies are validated using fixed version of Ensembl and all variants are annotated using Ensembl VEP.

Biosample

📖 Gentropy

The Experimental Factor Ontology, UBERON and Cell Ontology are composed as a meta-ontology to capture every tissue or cell type that could be described in a QTL study.

Each unique target–disease pair in the Open Targets Platform is defined as an association. For example, while there might be several pieces of evidence referring to CFTR and Cystic fibrosis from multiple sources, one single association contextualises all this information within the Platform. Also, since multiple pieces of evidence might refer to the same or similar associations, the Platform undertakes a series of steps to quantify their relative strength for a given association.

Ontological selection of evidence

The Platform associations aim to aggregate all evidence referring to the target and disease, but the complex phenotypic representation of the disease might sometimes cause different pieces of evidence to be annotated against slightly different levels of granularity of the disease.

For example, the association between inflammatory bowel disease and NOD2 is annotated with multiple pieces of evidence referring to these two terms specifically. The Platform refers to associations described by aggregated evidence between two specific terms in our data sources as direct associations. By default, direct associations are displayed in the web application when listing associated diseases or phenotypes with a target of interest.

However, evidence can sometimes be informative to discriminate targets in similar diseases or phenotypes. For example, when evaluating the inflammatory bowel disease and NOD2 association, other pieces of evidence describing the relationship between Crohn's disease and NOD2 might also be informative.

To approach this problem systematically, the Platform makes use of the properties of the disease ontology (EFO), to select all evidence referring to NOD2 in the context of inflammatory bowel disease or any of its ontology descendants (including Crohn's disease). This type of association is referred to in the platform as an indirect association. Indirect associations are displayed in the web application when displaying all the evidence for a target-disease association or listing associated targets with a disease or phenotype of interest.

The same calculations are applied to calculate the association scores, the only difference is the evidence included in the calculation.

Both direct and indirect associations can be queried using the GraphQL API or our data downloads page.

Association scores

Deciding what constitutes a strong association is open to interpretation. While some data sources can be more deterministic about the underlying causal evidence, others can occasionally point to targets indirectly linked to the disease. The Platform relies on a series of heuristics to maximise the transparency of the target and disease rankings.

By data source

The Platform's scoring by data source aims to take into account variations in how the data sources organise their evidence.

For example, some data sources are more stringent when it comes to defining what constitutes a single piece of evidence, whereas others rely on their internal evidence score to stratify the strength of the evidence they present.

Additionally, some data sources will capture the meaningful association in one single piece of evidence. In other data sources, the repetition of evidence increases our confidence in the association.

The scoring by data source aims to balance all these differences and provide a consensus view on the strength of the underlying evidence for a particular source.

For all cases, the Platform defines a data source association score by calculating a harmonic sum using the full vector of evidence scores as defined for each data source using the following the next steps:

1. The pieces of evidence are sorted in descending order and assigned an incremental value that indicates their position in the sorted list (the top-scoring item has a positional id of 1, the second has a positional id of 2, and so on).

2. The harmonic sum for each data source is then calculated by summing the result of dividing each evidence score by (positional id^2).

1. To ensure the result is between 0 and 1, the harmonic sum is normalised by dividing the result by the maximum theoretical harmonic sum, which is the one calculated using an infinite vector of ones. The platform derives this calculation (which approximates to 1.644) by using a vector of 1,000 ones.

evidence with score=1.0 -> positional id=1
evidence with score=0.9 -> positional id=2
evidence with score=0.8 -> positional id=3

harmonic sum score = 1.0/1^2 + 0.9/2^2 + 0.8/3^2

max. theoretical harmonic sum score = 1.0/1^2 + 1.0/2^2 + 1.0/3^2 + 1.0/4^2 + ... ~ 1.644
normalised harmonic sum score = harmonic sum score / max. theoretical harmonic sum score

By data type

The data type association score aims to capture the strength of the supporting evidence at the data type level (e.g. Genetic Associations). A second harmonic sum is calculated by using the vector of data source association scores weighted by the data source weights.

Overall

The overall association score aims to summarise all the aggregated evidence for a given target-disease association. The score is derived by calculating the harmonic sum of the association score by data source weighted by the data source weights, regardless of their data type categorisation. The algorithm to compute the scores is the same as the association by data source, resulting in a score between 0 and 1.

Data source weights

To calculate both data type and overall association scores, evidence is weighted using a factor that aims to calibrate the relevance of each data source relative to others. The default weights used in the web application can be modified by the user to adjust to different prioritisation strategies, both in the API and in the user interface through the "Advanced Option" tab from the new Associations on the Fly page.

Interpreting association scores

There are a few important considerations regarding association scores. As described above, association scores are a heuristic based on the availability of data. While scores are useful to rank lists of targets or diseases, they should not be interpreted as a confidence score for the target-disease association.

For example, under-studied diseases are unlikely to produce high-scoring targets due to the lack of available evidence. In such diseases, a relatively low-scoring target might still be the top-ranked target and potentially a very interesting lead from a therapeutic standpoint.

Similarly, not all associations with available target–disease evidence should be considered legitimate target–disease associations. Some of our data sources rely on predictions to assess the relationship between a target and a disease. Thus, they should be considered with caution and always take their relative support into consideration.

Fine-mapping is a statistical analysis technique used to pinpoint the specific genetic variant(s) most likely responsible for a trait association identified in a GWAS. The main result is a credible set (CS): the minimal list of variants with assigned posterior probabilities (PIPs) to be causal that form a predefined probability. The Platform uses 95% CS, meaning the CS has a 0.95 probability of containing the causal variant. The expected sum of PIPs for all variants in CSs have to be within the range of [0.95,1].

Open Targets Platform fine-mapping strategies (see more in Fine-mapping pipelines):

Clumping

📖 Gentropy

Clumping is a technique for selecting the most significant variants within a region or set of variants in LD, essentially collapsing signals into loci with an increased chance of containing a single causal variant for the phenotype. Three different methods are used for clumping:

Distance-based clumping

The method is based on an iterative procedure in which the variant with the strongest p-value is selected and all other variants within a predefined distance (radius) from this variant are clumped together. The procedure is repeated as long as there is at least one significant variant. The output of the method is a list of lead variants. There are two parameters: the distance to clump and the p-value significance threshold.

LD-based clumping

The method is applied to the results of the distance-based clumping (list of lead variants). It is again based on an iterative procedure in which the variant with the strongest p-value is selected and all other variants with high LD (r2>=0.5) are clumped together.

Locus breaker

The method consists of three steps:

1. In the first step, we perform the regular distance-based clumping.

2. In the second step, we filter the input summary statistics by the baseline p-value (default is 1e-5). We then clump variants that are closer to each other than the cutoff distance (default is 250,000 bp). Next, we filter clumps by having at least one variant with a p-value above the genome-wide significance threshold. At this stage, we define a list of loci consisting of information about the lead variant and the locus boundaries (the leftmost and rightmost variants in the clump). To each of the locus boundaries we subtract/add (to the left/right boundaries respectively) the flanking distance (the default is 100,000 bp) to avoid situations where the locus size consisting of only one lead variant is 0.

3. In the third step we select the loci with the size greater than the specified large locus size threshold (1,500,000 bp by default) and 'break' each of the large loci using the lead variants from the distance-based clump that lie within the boundaries of the large locus. For each of the distance-based lead variants, we assigned the boundaries as +/-half the size of the large locus size threshold. Thus, each large locus is divided by several overlapping loci of the size of the large locus size threshold. The small loci are unaffected by the splitting.

The general procedure results in a list that contains information about the lead variant and the locus boundaries (the most left and right variants in the cluster), and the largest locus size doesn't exceed the large locus size threshold. The boundaries of the locus are then used to define the region for fine-mapping and LD matrix ingestion. The reason for using this procedure is because locus breaker results in smaller locus sizes on average compared to those defined by window-based clumping.

Acknowledgements: We are grateful to our colleagues at Human Technopole, Sodbo Sharapov and Nicola Pirastu, for advice on this method.

PICS fine-mapping

📖 Gentropy

The PICS algorithm was originally implemented in Farh et al. (2015) investigating the fine-mapping of causal autoimmune disease variants. It is a method to fine-map the most likely causal variants associated with a trait or disease within a haplotype. The algorithm is based on the calculation of the Posterior Inclusion Probabilities (PIP) of tag variants linked to the lead variant by LD within the target population. Only five populations are used for PICS fine-mapping: African-American (AFR), American Admixed/Latino (AMR), East Asian (EAS), Finnish (FIN), and Non-Finnish European (NFE). The calculation is performed by using all the proxy variants where r2 ≥ 0.5 and the default parameter k=6.4 as reported in the original paper.

SuSiE-inf fine-mapping

📖 Gentropy

If both summary statistics and high-precision LD are available for the locus, we use SuSiE-inf fine-mapping (see Cui et al. (2024) for more details). This method is a generalisation of the original SuSiE, allowing modelling of infinitesimal effects alongside fewer larger causal effects.

SuSiE-inf has two approaches for updating estimates of the variance components: Method of Moments and Maximum Likelihood Estimator ('MoM' / 'MLE'). The function takes an array of Z-scores and a numpy array matrix of variant LD to perform fine-mapping. There is a boolean option “est_tausq” that enables the estimation of infinitesimal effects. If it is disabled, it performs fine-mapping similar to the original SuSiE method. We use SuSiE-inf only with the combination of pan-UK Biobank LD matrices for three ancestries: NFE, CSA and AFR.

Fine-mapping pipelines

Different pipeline strategies have been defined for different sources based on a combination of the above methodologies:

Clumping and fine-mapping of GCSS and GCCA PICS

Distance and LD-based clumping is applied to all available GCSS and GCCA. P-value significance is defined at <=1e-8 threshold and distance-based clumping is performed using a 500,000 bp radius. LD-clumping is performed on the same window using the LD dataset from gnomAD v2.1.1. PICS fine-mapping is then performed using the same LD information and credible sets are defined based on 95% inclusion probability. In the case of multiple ancestry studies, we used Non-Finnish Europeans (NFE) if it was in the list of ancestries, or major ancestry otherwise. In the case the study's ancestry is not in the list available in LD index ancestries, no fine-mapping results are produced. If the lead variant is not present in the LD-index, the credible set contains only that variant and it's flagged with the label Variant not found in LD reference. All resulting credible sets objects were additionally flagged with the label Study locus fine-mapped without in-sample LD reference.

Clumping and fine-mapping of GCSS SuSiE

GWAS-significant study-locus are identified using a p-value threshold <=1e-8 and the locus-breaker method. SuSiE-inf fine-mapping was applied on all study-loci following the next criteria:

1. Major ancestry for the study is NFE, AFR, CSA or EAS. If the major ancestry was EAS, we used CSA instead

2. The study type is "gwas"

3. The study has no analysis flags except "metabolite"

4. Study has no quality control flags

5. The locus didn't overlap with the MHC region. We didn’t exclude X or Y chromosomes if they were presented in GWAS

6. The number of variants in the locus after overlapping with the LD matrix was in the range [100, 15,000]

For all eligible study loci, the SuSiE-inf method was applied without estimation of infinitesimal effects (equivalent to the classical SuSiE method) using pan-UKBB LD matrices. Filtering of the resulting 95% CSs is performed based on CS log(BF)<=2, minimum R2 purity <=0.25 and lead variant p-value>=1e-5. Additionally, the pairwise r^2 between lead variants within the locus is calculated, removing the less significant of the CSs if the r2>=0.8. As the locus breaker procedure can create overlapping loci, we can obtain the redundant credible sets within the same study. Thus, if two CSs from different loci within a study had the same lead variant, we removed one of them, leaving the CS with the largest CS log(BF). All resulting credible sets from this pipeline are flagged with the label Study locus fine-mapped without in-sample LD reference.

UKBB-PPP clumping and fine-mapping

The UKBB-PPP summary statistics are clustered into locus using the locus breaker method using a p-value significance threshold <=1.7e-11. The resulting study-loci are fine-mapped using the SuSiE-inf method with Pan-UKBB LD matrices for the EUR population. The resulting 95% CSs are filtered similar to GCSS SuSiE pipeline described above. All resulting StudyLocus objects were additionally flagged with the label Study locus fine-mapped without in-sample LD reference.

Overview

A credible set is a set of genetic variants near a genetic association signal that is predicted, with a specific probability, to include the causal variant for that signal. The results of the fine-mapping analysis determine this, assigning each variant in the region a posterior probability of being causal when considering the observed statistics and the population structure. The variants covering the top 95% likelihood of containing the causal variant define the credible sets in the Platform.

A credible set results from statistical analysis on a specific locus in a study. As a consequence, all credible sets are defined as:

• Study in which the association is reported

• Lead variant - Variant with the highest posterior probability in the credible set

• Fine-mapping method and statistics

The Platform contains every credible set resulting from fine-mapping all our sources after applying certain exclusion criteria.

Credible set exclusion criteria

Credible sets fulfilling any of the next rules are excluded from the Platform:

1. The lead variant is within the MHC region (chr6:25726063-33400556)

2. The credible set is not in a valid study

3. There is another fine-mapped SuSiE credible set from the same region and study

4. Being a GWAS catalog fine-mapped top-hit, there is a GWAS catalog fine-mapped credible set from summary statistics for the same region and study

5. The lead variant is reported in an invalid chromosome (1:22, X, Y, XY, MT)

6. The sum of PIPs in the credible is not within the [0.95,1] range

Credible set confidence

Credible sets are categorised based on the fine-mapping confidence, derived from the available association data, fine-mapping framework and availability of linkage disequilibrium (LD) information for the specific study/population . The categories in descending order of confidence are:

• SuSiE or SuSiE-inf fine-mapped credible set with in-sample LD

• SuSiE or SuSiE-inf fine-mapped credible set with out-of-sample LD

• PICS fine-mapped credible set extracted from summary statistics

• PICS fine-mapped credible set based on reported curated association

• Unknown confidence

The confidence values are symbolised by 0 to 4 stars on the UI. It is important to note that this confidence does not reflect the strength of the association or the effect size.

Credible set variants

All variants in the credible set are annotated with:

• P-value. Unconditioned p-value from the study when available.

• Beta. Corresponds to SuSiE mu (SuSiE fine-mapped GWAS Catalog, UKBB-PPP and FinnGen) or beta (PICS fine-mapped GWAS Catalog and SuSiE fine-mapped eQTL Catalog)

• Standard error. Only available for PICS-fine-mapped credible sets.

• LD (r^2). Linkage-disequilibrium information. Only available for PICS-fine-mapped credible sets.

• Posterior probability. Posterior inclusion probability (PIP) of variant being causal after fine-mapping.

• log(BF). The logarithm of the Bayes factor. Only available for SuSiE credible sets.

• Predicted consequence. The most severe consequence is across all overlapping canonical transcripts, as reported by the Ensembl VEP.

Locus-to-Gene (L2G)

Machine-learning prioritisation of likely causal genes based on available features. L2G integrates multiple features to predict what's the most likely causal gene in the neighbourhood of the observed association. All predictions for protein-coding genes with a score above 0.05 are displayed. See Locus-to-Gene section for a description of the methodology.

Explaining L2G predictions

Using the SHAP (SHapley Additive exPlanations) library, we have extracted feature importance values for all L2G predictions. Shapley values provide a principled approach based on game theory to explain the contribution of individual features or groups of features, revealing how each group influences the final L2G score. These contributions are approximated to be additive, meaning the sum of the Shapley values for all feature groups equals the total L2G score or reasonably close to it.

We have aggregated the Shapley values into the main feature groups to understand their relative importance.

The base value represents the baseline before any feature-specific information is considered, and is therefore equivalent for all genes and credible sets.

This approach helps to identify which types of evidence (e.g. distance, colocalisation or functional impact) are most influential for a given locus-gene association. In addition, we can visualise the individual contribution of each feature within the group. As the features within the group are highly correlated, the individual values are less interpretable than the group contribution.

A full description of all features is available here.

Source: Open Targets

Colocalisation

Credible sets are compared against other credible sets to find overlapping signals. Two overlapping credible sets are those that share at least one variant in the set. The Platform contains all the overlaps between all GWAS vs all GWAS and all GWAS vs all molQTL studies.

For the overlapping pairs of credible sets, estimates for two co-localisation methods are computed based on the type of credible set:

• COLOC - for overlapping SuSiE fine-mapped credible sets

• eCAVIAR - for all overlapping credible sets

Source: Open Targets

Colocalisation directionality

A directionality assessment is included in the colocalisation analysis to help the user interpret the relationship between the two overlapping credible sets.

For every overlapping variant in a pair of overlapping credible sets, the sign of the ratio between both beta estimates are calculated (+1 or -1) indicating the individual variant directionality. The average of these signs across all overlapping variants in both credible sets is used to estimate the sign estimates. When the average approximates to +1, the two credible sets are assessed to share the same directionality. If the average approximates to -1, the two credible sets are interpreted to have opposite directionality. In any other case, the assessment is declared inconclusive (N/A).

Source: Open Targets

Pairs of credible sets overlapping at least one variant are subject to additional analysis to infer the likelihood of them sharing the same causal variant. The Platform compares all GWAS vs all GWAS credible sets and all GWAS vs all molQTL credible sets. For overlapping credible sets, co-localisation metrics are estimated using the following methods:

COLOC

• H0: No association with either trait

• H1: Association with trait 1, not with trait 2

• H2: Association with trait 2, not with trait 1

• H3: Association with trait 1 and trait 2, two independent SNPs

• H4: Association with trait 1 and trait 2, one shared SNP

To reduce potential false positives in small overlaps, if the size of the overlap is less than 5 SNPs, we check whether the maximum product of the PIPs for at least one of the overlapping SNP pairs is greater than 0.01. Otherwise, the overlap is excluded from the COLOC analysis.

eCAVIAR

Gentropy is an open-source Python package to facilitate the interpretation and analysis of GWAS and functional genomic studies for target identification. The Platform leverages Gentropy to perform post-GWAS analysis and derive the evidence and datasets for web portal visualisation.

Gentropy provides a set of data models, data ingestion methods and statistical analysis organised in discrete steps to maximise re-usability. Gentropy is designed with scalability in mind, so it's suitable for small dedicated analysis but also for the high-performing orchestrated tasks required to generate the data-lake that populates the Open Targets Platform.

Target–disease evidence

Every event or set of events pinpointing a target as a potential causal gene or protein for a disease represents the unit of information, most often referred to as evidence. Within the Open Targets Platform, a series of pipelines ensure information is retrieved from its sources and standardised in a way that can be immediately applied to answer drug development queries.

All evidence is mapped to the reference target entity identifier (Ensembl gene) and disease or phenotype identifier (experimental factor ontology, EFO), as well as other reference controlled vocabularies and ontologies when appropriate. Evidence is also reviewed to minimise the presence of duplicates within the same data source.

Data sources are also grouped into bigger categories abstracting the type of evidence they predominantly capture. In the platform, these categories are usually referred to as data types, as opposed to the individual resource data referred to as data sources.

The Open Targets Platform provides a scoring framework for each data source to contextualise the relative importance of each piece of evidence. This score will be more relevant when understanding the association scoring in later sections.

Evidence data sources

GWAS associations

The GWAS associations data source aggregates target-disease relationships supported by significant genome-wide associations (GWAS) in the context of other functional genomics data.

Datatype: Genetic associations

Gene Burden

Gene burden data comprises gene–phenotype relationships observed in gene-level association tests using rare variant collapsing analyses. The Platform integrates burden tests carried out by several sources:

• REGENERON (Backman et al., 2021), a whole-exome sequencing analysis of individuals from the UK Biobank.

• AstraZeneca PheWAS Portal (Wang et al., 2021), a whole-exome sequencing analysis of individuals from the UK Biobank.

• Genebass (Karczewski et al., 2022): Gene-based Association Summary Statistics (Genebass), a whole-exome sequencing analysis of individuals from the UK Biobank.

• The results of whole-exome and whole-genome sequencing analysis based on the SPARK cohort bring evidence of novel targets implicated in autism spectrum disorder (Zhou et al., 2022).

• The SCHEMA consortium (Singh et al., 2022), a whole-exome sequencing analysis of individuals with schizophrenia.

• The Epi25 collaborative (Epi25 Collaborative, 2019), a whole-exome sequencing analysis of individuals with epilepsy.

• The Autism Sequencing Consortium (Satterstrom et al., 2020), a whole-exome sequencing analysis of individuals with autism spectrum disorder.

• The results of an Open Targets project (Bomba et al., 2022), a whole-exome sequencing analysis of individuals from the INTERVAL cohort testing for associations between rare coding variants and blood metabolites.

• The results of a pan-ancestry whole-exome sequencing analysis identify relevant genes associated with fat distribution (Akbari et al., 2022).

• The results of whole-exome and whole-genome sequencing analysis on Parkinson disease and promoted by the AMP-PD initiative, and other collaborators (Makarious et al., 2022).

• The results of gene-based analyses of rare variants and circulating metabolic biomarkers relevant to cardiovascular disease (Riveros-McKay et al., 2020).

• The results of rare coding variant analyses from whole exome sequencing of Black South African men to identify genes significantly associated with prostate cancer (Soh et al., 2023)

These associations are a result of collapsing rare variants in a gene into a single burden statistic and regress the phenotype on the burden statistic to test for the combined effects of all rare variants in that gene. The different collapsing methods inform about the filters used to select the set of qualifying variants, mostly based on their pathogenicity and frequency in the population.

Datatype: Genetic associations

Evidence scoring: Scaled p-value from 0.25 (p = 1e-7) to 1 (p < 1e-17).

Direction of Effect assessment:

ClinVar

ClinVar is an NIH public archive of reports of the relationships among human variations and phenotypes, with supporting evidence. The ClinVar data source in the Open Targets Platform captures the subset of ClinVar that refers to germline variants (as opposed to somatic variants). Each evidence in the platform aims to capture an individual RCV record in ClinVar.

Information on variants is covered extensively for both single point and structural variants. When available, genomic coordinates are reported with RS numbers, or by following the CHROM_POS_REF_ALT and HGVS notations.

Datatype: Genetic associations

Evidence scoring: ClinVar evidence is scored in a 2-step process. In Step 1, a score is assigned to every piece of evidence based on the clinical significance:

In Step 2, the score is modulated based on the ClinVar review status:

Direction of Effect assessment:

Genomics England (GEL) PanelApp

The Genomics England PanelApp is a knowledge base that combines crowdsourced expertise with curation to provide gene–disease relationships. Virtual gene panels related to human disorders are reviewed by experts within the clinical and scientific community to support the interpretation of genomes within the 100,000 Genomes Project. Within a panel, genes are rated based on the level of evidence supporting the association with the phenotypes identified by the panel. Genes are then classified according to a traffic light system with red/stop, amber/pause, and green/go classifications. To receive a green rating (diagnostic-grade) on a version 1+ panel, the gene requires "evidence from 3 or more unrelated families or from 2-3 unrelated families where there is strong additional functional data" and "genes that do not meet these criteria are rated as Amber (borderline) or Red (low level of evidence)."

Data type: Genetic associations

Evidence scoring: Based on Genomics England gene rating:

Gene2Phenotype

G2P evidence in the Platform is the result of any target-disease curation by any of the expert panels.

Data type: Genetic associations

Evidence scoring:

Direction of Effect assessment:

UniProt literature

The Universal Protein Resource (UniProt) provides a large compendium of sequence and functional information at the protein level. As part of their functional annotation effort, UniProt curators also annotate proteins with publications supporting their involvement on pathogenic processes.

All publications supporting a given target disease relationship are aggregated into one single Platform evidence.

Data type: Genetic associations

Evidence scoring:

UniProt variants

The Universal Protein Resource (UniProt) also curate variants supported by publications that are known to alter protein function on disease. Curated mutations are predominantly protein coding or in regulatory regions clearly associated with the causal protein.

All publications supporting a given variant in connection with a disease constitute individual evidence. All supporting publications are aggregated within the same evidence.

Data type: Genetic associations

Evidence scoring:

Orphanet

Orphanet is an international network that offers a range of resources to improve the understanding of rare disorders of genetic origin. These resources include an inventory of rare disease and gene associations, classification of the gene–disease relationship, information on the kind of mutation, and supporting publication references.

Data type: Genetic associations

Evidence scoring:

Direction of Effect assessment:

ClinGen

The Clinical Genome Resource (ClinGen) Gene–Disease Validity Curation aims to evaluate the strength of evidence supporting or refuting a claim that variation in a particular gene causes a particular disease. ClinGen provides a framework of guidelines to assess clinical validity in a semi-quantitative manner allowing curators to classify the validity of given gene–disease pair.

All gene–disease pairs mapped to EFO constitute individual evidence in the Platform.

Data type: Genetic associations

Evidence scoring:

ChEMBL

EMBL-EBI's ChEMBL is a manually curated database of bioactive molecules with drug-like properties, either approved for marketing by the U.S Food and Drug Administration (FDA), or clinical candidates. ChEMBL also captures information regarding the drug molecule indications, as well as their curated pharmacological target.

In the Platform, ChEMBL evidence represents any target–disease relationship that can be explained by an approved or clinical candidate drug, targeting the gene product and indicated for the disease. Independent studies are treated as individual evidence.

To provide additional context, we integrate a machine learning-based analysis of the reasons why a clinical trial has ended earlier than scheduled. This sorts the stop reasons into a set of 17 classes which include negative, neutral, and positive reasons. This information is available when hovering on the tooltip of the Source column.

The 17 classes are: Another Study, Business or Administrative, Negative, Study Design, Invalid Reason, Ethical Reason, Insufficient Data, Insufficient Enrolment, Study Staff Moved, Endpoint Met, Regulatory, Logistics or Resources, Safety and Side Effects, No Context, Success, Interim Analysis, and Covid 19.

Data type: Drugs

Evidence scoring: ChEMBL evidence is scored in a 2-step process. In Step 1, a score is assigned to every piece of evidence based on the clinical precedence:

In Step 2, for those clinical trials that have stopped early, the score is down-weighted based on the classification of the reason to stop. In this way, less importance is attributed to evidence of studies that have been stopped due to negative outcomes or safety concerns:

Direction of Effect assessment:

Reactome

The Reactome database manually curates and identifies reaction pathways that are affected by a disease. Reactome annotation includes information regarding the causal target–disease link either being a protein coding mutation or an altered expression.

In the Platform, any mutation or altered expression event affecting a different reaction is captured in a different target–disease evidence.

Data type: Pathways & systems biology

Evidence scoring: All manually curated evidence in Reactome has a score of 1.

CRISPR screens

One of the most powerful approaches to uncover gene function is the experimental perturbation of genes followed by the observation of related phenotypes. The perturbation of gene function in human cells has been greatly facilitated by developments in CRISPR technology.

We have linked cell types to diseases, meaning these diseases are often characterised with abnormal phenotypes in these cell types — hence the association. If knocking out a gene causes significant perturbation in the cell type, it might indicate a potential targeting strategy in the disease.

Data Type: Pathways & systems biology

Evidence Scoring: The Platform uses the linearised CRISPRbrain's assessment of statistical significance to assign a score, including hits from both the upper and lower end of the distribution

Project Score

Project Score is a Wellcome Sanger Institute resource that aims to identify dependencies in cancer cell lines to guide precision medicine. The project combines gene fitness effects derived from whole-genome CRISPR-Cas9 synthetic lethality screenings with tractability data, genomic biomarkers and various target annotation enabling a systematic prioritisation of potential targets. The resulting inferences are then mapped from the cancer cell lines in which the experiment is performed to their corresponding tumours.

In the Platform, any Project Score prioritised target with priority score reaching 36.0 is included as independent evidence; however, pan-cancer dependencies are excluded from the integration.

Data type: Pathways & systems biology

Evidence scoring: Project Score priority score divided by 100

SLAPenrich

In the Platform, each pathway significantly enriched in tumour-occurring mutations constitutes an individual piece of evidence.

Data type: Pathways & systems biology

Evidence scoring: Scaled enrichment p-value from 0.5 (p = 1e-4) to 1 (p<1e-14).

Gene signatures

The Platform also provides information about key driver genes for specific diseases that have been curated from Systems Biology analysis. These publications present different disease gene signatures as potential key drivers or key regulators causing disease.

Data type: Pathways & systems biology

Evidence scoring: Scoring depends on whether the original data contains or not a score:

• p-values and rank-based scores are normalised to the 0.5 - 1 range

• If there is no score a fixed value of 0.5 is used

PROGENy

In the Platform, a PROGENy evidence is defined as any significantly regulated sample-level pathway activities inferred from matched normal vs. tumour samples.

Data type: Pathways & systems biology

Evidence scoring: Scaled p-value from 0.5 (p = 1e-4) to 1 (p<1e-14).

Expression Atlas

The EMBL-EBI Expression Atlas provides a differential expression pipeline aiming to identify genes that are differentially expressed in disease vs control samples. Only contrasts from studies with enough replicates and minimum quality criteria are included in the processing.

In a given contrast, to consider a gene significantly regulated in a contrast, all the following rules are required:

• Absolute log2 fold change > 1

• Adjusted p-value <= 0.05

• Maximum significant genes probes per contrast = 1000

In the Platform, each contrast from independent studies capturing differentially regulated genes constitutes independent evidence.

Data type: RNA expression

Evidence scoring: ExpressionAtlas scoring is the result of the product of:

• Scaled p-value from 0 (p = 1) to 1 (p<1e-10)

• Absolute log2 fold change divided by 10

• Percentile rank divided by 100

Cancer Gene Census

In the Platform, CGC evidence is aggregated at the target–disease level to provide a summary of all curated evidence supporting the involvement of a target with a particular cancer type.

Data type: Somatic mutations

IntOGen

In the Platform, independent target–disease evidence are defined as any significant driver gene detected in any individual cohort. Information regarding the individual driver methods is also provided within each evidence.

Data type: Somatic mutations

ClinVar (somatic)

ClinVar is an NIH public archive of reports of the relationships among human variations and phenotypes, with supporting evidence. The ClinVar (somatic) data source in the Open Targets Platform captures the subset of ClinVar that refers to somatic variants (as opposed to germline variants).

Information on variants is covered extensively for both single point and structural variants. When available, genomic coordinates are reported with RS numbers, or by following the CHROM_POS_REF_ALT and HGVS notations.

Each evidence in the Platform aims to capture an individual RCV record in ClinVar.

Datatype: Somatic mutations

Evidence scoring: ClinVar evidence is scored in a 2-step process. In Step 1, a score is assigned to every piece of evidence based on the clinical significance:

In Step 2, scored is modulated based on the ClinVar review status:

Direction of Effect assessment:

Europe PMC

The EMBL-EBI's Europe PMC enables access to a worldwide collection of life science publications and preprints from trusted sources. The Europe PMC data source aims to identify target–disease co-occurrences in the literature and provide an assessment on the confidence of the relationship. This pipeline uses deep-learning based Named Entity Recognition (NER) to identify gene/proteins and diseases when mentioned in the text, to later normalise them to the target or disease/phenotype entities in the Platform. All co-occurrences of both types of entities in the same sentence are considered evidence.

In the Platform, a piece of Europe PMC evidence is the result of aggregating all co-occurrences of the same target and disease within the same publication.

Data type: Text mining

IMPC

The genotype–phenotype associations made available by the International Mouse Phenotypes Consortium (IMPC) are used to identify models of human disease based on phenotypic similarity scores.

The Wellcome Sanger Institute PhenoDigm is an algorithm aimed at capturing the similarity between a knockout mouse and the clinical manifestations (phenotype) of a human disease. The premise is that if a gene knock-out causes an equivalent phenotype in mouse, the human counterpart is likely to be related with the cause of the disease.

It uses a semantic approach to map between clinical features observed in humans and mouse phenotype annotations. The phenotypic effects in mice are then mapped to phenotypes associated with human diseases. The matches are identified and a similarity score between a mouse model and a human disease is computed.

Data type: Animal models

Direction of Effect assessment:

Cancer Biomarkers

One of the aims of the Cancer Genome Interpreter is to identify how variations in the tumour genome may influence its response to anti-cancer therapies. The Cancer Biomarkers database features biomarkers of drug sensitivity, resistance, and toxicity for drugs targeting specific targets in cancer, curated by clinical and scientific experts in precision oncology, and classified by cancer type.

Data type: Pathways & systems biology

Evidence scoring: All manually curated evidence in Cancer Biomarkers has a score of 1

To support more complex and systematic queries, we provide all datasets as data downloads.

A list of all datasets is available in the Platform Data Downloads page.

All Platform datasets are available as a distributed collection of data. This implies that for each dataset, there will be a directory with a list of partitioned files. Currently, we produce our datasets in Parquet. This formats allow us to expose nested information in a machine-readable way. Next we describe how to download, access and query this information in a step-by-step guide.

Archive datasets, as well as input files and other secondary products are also made available in the FTP server and Google Cloud Platform.

Download

Below is a walkthrough on how to download the disease dataset from the 25.03 release in Parquet format using different approaches.

We recommend using lftp with a command line client, and when using tools like wget, curl, etc., use https:// rather than ftp://

Using rsync

rsync is a command line tool for efficiently transferring and synchronising files between a computer and an external hard drive.

rsync -rpltvz --delete rsync.ebi.ac.uk::pub/databases/opentargets/platform/25.03/output/disease .

Using wget

wget is a command line tool that retrieves content from web servers and widely available in Unix systems.

wget --recursive --no-parent --no-host-directories --cut-dirs 8 \
https://ftp.ebi.ac.uk/pub/databases/opentargets/platform/25.03/output/disease

Using Google Cloud Platform (paywalled after 1TB)

Users with Google Cloud Platform account can download the datasets through the Google Cloud Console or using gsutil command-line tool.

gsutil -m cp -r gs://open-targets-data-releases/25.03/output/disease

Other ways to access data

If you are using a non-Linux or non-Unix machine (e.g. Windows), you can access our FTP service using an FTP client like FileZilla or the Windows ftp command. For more information, including tips and workarounds, see the Community Windows ftp thread.

Accessing and querying datasets

To read the information available in the partitioned datasets, there is no need to manipulate or concatenate files. Datasets can be read directly using the dataset path.

The next scripts provide a proof-of-concept example using the ClinVar evidence provided by the European Variation Archive. The next scripts show how to:

• Read a dataset

• Explore the schema of the dataset

• Select a subset of information (columns)

• Display the information

First of all the dataset needs to be downloaded as described in the previous section. For simplicity, only EVA evidence is downloaded, but all evidence can be downloaded at once using the same approach.

gsutil -m cp -r gs://open-targets-data-releases/25.03/output/evidence/sourceId=eva

The next scripts make use of Apache Spark (PySpark or Sparklyr) to read and query the dataset using modern functional programming approaches. These packages need to be installed in their respective environments.

The next query only displays 6 fields of the ClinVar evidence but there are other non-null values available. The schema is the best way to explore what's available and query the most relevant information. All Platform evidence share the same schema, so there will be a long list of fields that might not be informative for ClinVar but will be relevant if trying to query other data sources.

Dealing with nested information can sometimes be tedious. The Platform aims to minimise the nestiness of the data, however some level of structure is sometimes required. Spark provides a series of functions to deal with complex nested information. The scripts provide an example on how the clinicalSignificances array is flattened using the explode function.

Once loaded into Python or R, the user can decide to continue using Spark, write the output to a file or use alternative libraries to process the information (e.g. pandas, tidyverse, etc.).

from pyspark import SparkConf
from pyspark.sql import SparkSession
import pyspark.sql.functions as F

# path to ClinVar (EVA) evidence dataset 
# directory stored on your local machine
evidencePath = "local directory path - e.g. /User/downloads/sourceId=eva"

# establish spark connection
spark = (
    SparkSession.builder
    .master('local[*]')
    .getOrCreate()
)

# read evidence dataset
evd = spark.read.parquet(evidencePath)

# Browse the evidence schema
evd.printSchema()

# select fields of interest
evdSelect = (evd
 .select("targetId",
         "diseaseId",
         "variantRsId",
         "studyId",
         F.explode("clinicalSignificances").alias("cs"),
         "confidence")
 )
 evdSelect.show()

# +---------------+--------------+-----------+------------+--------------------+--------------------+
# |       targetId|     diseaseId|variantRsId|     studyId|                  cs|          confidence|
# +---------------+--------------+-----------+------------+--------------------+--------------------+
# |ENSG00000153201|Orphanet_88619|rs773278648|RCV001042548|uncertain signifi...|criteria provided...|
# |ENSG00000115718|  Orphanet_745|       null|RCV001134697|uncertain signifi...|criteria provided...|
# |ENSG00000107147|    HP_0001250|rs539139475|RCV000720408|       likely benign|criteria provided...|
# |ENSG00000175426|Orphanet_71528|rs142567487|RCV000292648|uncertain signifi...|criteria provided...|
# |ENSG00000169174|   EFO_0004911|rs563024336|RCV000375546|uncertain signifi...|criteria provided...|
# |ENSG00000140521|  Orphanet_298|rs376306906|RCV000763992|uncertain signifi...|criteria provided...|
# |ENSG00000134982|   EFO_0005842| rs74627407|RCV000073743|               other|no assertion crit...|
# |ENSG00000187498| MONDO_0008289|rs146288748|RCV001111533|uncertain signifi...|criteria provided...|
# |ENSG00000116688|Orphanet_64749|rs119103265|RCV000857104|uncertain signifi...|no assertion crit...|
# |ENSG00000133812|Orphanet_99956|rs562275980|RCV000367609|uncertain signifi...|criteria provided...|
# +---------------+--------------+-----------+------------+--------------------+--------------------+
# only showing top 10 rows

# Convert to a Pandas Dataframe
evdSelect.toPandas()

library(dplyr)
library(sparklyr)
library(sparklyr.nested)

## path to ClinVar (EVA) evidence dataset 
## directory stored on your local machine
evidencePath <- "local directory path - e.g. /User/downloads/sourceId=eva"

## establish connection
sc <- spark_connect(master = "local")

## read evidence dataset
evd <- spark_read_parquet(sc,
                          path = evidencePath)
## Browse the evidence schema
columns <- evd %>%
  sdf_schema() %>%
  lapply(function(x) do.call(tibble, x)) %>%
  bind_rows()

## select fields of interest
evdSelect <- evd %>%
  select(targetId,
         diseaseId,
         variantRsId,
         studyId,
         clinicalSignificances,
         confidence) %>%
  sdf_explode(clinicalSignificances)

##  # Source: spark<?> [?? x 6]
##    targetId   diseaseId   variantRsId studyId  clinicalSignific… confidence     
##    <chr>      <chr>       <chr>       <chr>    <chr>             <chr>          
##  1 ENSG00000… Orphanet_8… rs773278648 RCV0010… uncertain signif… criteria provi…
##  2 ENSG00000… Orphanet_7… NA          RCV0011… uncertain signif… criteria provi…
##  3 ENSG00000… HP_0001250  rs539139475 RCV0007… likely benign     criteria provi…
##  4 ENSG00000… Orphanet_7… rs142567487 RCV0002… uncertain signif… criteria provi…
##  5 ENSG00000… EFO_0004911 rs563024336 RCV0003… uncertain signif… criteria provi…
##  6 ENSG00000… Orphanet_2… rs376306906 RCV0007… uncertain signif… criteria provi…
##  7 ENSG00000… EFO_0005842 rs74627407  RCV0000… other             no assertion c…
##  8 ENSG00000… MONDO_0008… rs146288748 RCV0011… uncertain signif… criteria provi…
##  9 ENSG00000… Orphanet_6… rs119103265 RCV0008… uncertain signif… no assertion c…
## 10 ENSG00000… Orphanet_9… rs562275980 RCV0003… uncertain signif… criteria provi…
## # … with more rows

# Convert to a dplyr tibble
evdSelect %>%
  collect()

Tutorials and how-to guides

For more information on how to access and work with our data downloads and example scripts based on actual use cases and research questions, check out the Open Targets Community.

To support systematic identification and prioritisation of drug targets, we are committed to making our data open and available in a variety of formats to support academic research purposes and commercial activities. For more information, see our Licence documentation.

For queries about a single entity or target-disease association, we recommend that you use:

• Our intuitive web interface with data tables and data visualisations that can be downloaded/exported in multiple formats, including TSV and JSON

• Our robust GraphQL API with endpoints that can be accessed with the programming language of your choice and an interactive playground where you can try out sample queries

For more complex, systematic queries, we recommend that you use:

• Our comprehensive set of data downloads available via FTP, Google Cloud, and Microsoft Azure Open Datasets.

• Our Google BigQuery instance that supports SQL-like queries and allows you to export data to your own Google Cloud Storage bucket. This data is available as a BigQuery Public Dataset.

If you use our data in your research or commercial product, please cite our latest publication.

New export functionality

We have designed and developed a new export functionality for the Associations On the Fly/Target Prioritisation pages, allowing users to download:

• Entire dataset view (default status)

• Customised dataset view including custom controls changes, subset of data types (aggregations) and/or data from pinned targets only

• TSV and JSON formats

The new target prioritisation page can be accessed by clicking onto the Target prioritisation factors tab from the Associations on the Fly page when searching for targets associated with a disease or phenotype.

The view focuses on displaying target-specific properties in a disease agnostic way, which have been aggregated into four main sections—Precedence, Tractability, Doability, Safety—and individually scored by the Open Targets team.

A "traffic light" system has been designed to visually inform on target prioritisation, with the aim to facilitate target recommendations. Using a colour scale, green indicates potentially positive attributes and red indicates potentially negative attributes, providing information to help users assess the targets for further prioritisation or deprioritisation, respectively.

Precedence section

Target in clinic

Definition: Gene is targeted by available drugs in any clinical phase for any indication.

Source of Data: Platform Known Drugs widget (ChEMBL)

Scoring: Maximum clinical trial phase the target has been reported for, independently of the disease. Phases range from 0 to IV (corresponding to values of 0, 0.25, 0.5, 0.75 and 1 in the tool scores).

Tractability section

Membrane Protein

Definition: Target is annotated to be located in the cell membrane.

Source of Data: Platform Subcellular location widget [HPA (Human Protein Atlas) and UniProt]

Scoring:

• 1 = Protein target is located (at least) in the cell or plasma membrane.

• 0 = Protein target is not located in the cell membrane but some location information is accessible.

• NA = No location information available.

Secreted protein

Definition: Target is secreted or predicted to be secreted.

Source of Data: Platform Subcellular location widget [HPA (Human Protein Atlas) and UniProt]

Scoring:

• 1 = Protein target is (at least) secreted or predicted to be secreted.

• 0 = Not secreted but with location information.

• NA = No location information available.

Ligand binder

Definition: Target binds at least one High-Quality Ligand according to ChEMBL tractability bucket.

Source of Data: Platform tractability widget (Open Targets tractability)

Scoring:

• 1 = Target has a high-quality ligand reported.

• 0 = Target does not have high-quality ligand reported.

• NA = No information available.

Small molecule binder

Definition: Target has been co-crystallised with a small molecule, reported in the Protein Data Bank.

Source of Data: Platform tractability widget (Open Targets tractability)

Scoring:

• 1 = Target has a small molecule reported.

• 0 = Target does not have a small molecule reported.

• NA = No information available.

Predicted pockets

Definition: Target has a DrugEBIlity score equal or above 0.7, which is predictive of harbouring a high-quality pocket.

Source of Data: Platform tractability widget (Open Targets tractability)

Scoring:

• 1 = Target contains a high-quality predicted pocket.

• 0 = Target does not have a high-quality predicted pocket.

• NA= No information available.

Doability section

Models, tools and/or reagents that allow target assessment in preclinical settings to enable exploration of a given target

Mouse ortholog identity

Definition: Mouse orthologs maximum identity percentage. A mouse harbouring an ortholog for the target of interest could be useful for in vivo assaying.

Source of Data: Platform comparative genomics widget (Ensembl Compara)

Scoring:

From 0 to 1 are linearly scored those targets with at least one ortholog in mice harbouring at least 80% with the target.

• 1 = There is at least one gene in mice that contains a sequence with a 100% of identity with the target.

• 0 = There are no genes in mice containing a sequence with at least 80% of identity with the target.

• NA = No ortholog information.

Chemical probes

Definition: Target has high quality chemical probes.

Chemical probes are small molecules acting as chemical modulators, binding reversibly to the target.

Source of Data: Platform Chemical probes widget (Probes & Drugs)

Scoring:

• 1 = Target has high-quality chemical probes.

• 0 = Target does not have high-quality chemical probes.

• NA = No information available.

Safety section

Genetic constraint

Definition: Genest that are important for human physiology are seen to be depleted of deleterious variants. The Genome Aggregation Database (gnomAD) has developed a continuous measurement of intolerance to loss of function (LoF) variants per gene, based on observed/expected LoF variant analysis. As recommended by gnomAD and implemented in the Open Targets platform, the rank of genes regarding their loss-of-function observed/expected upper bound fraction (LOEUF) metric is used (LOEUF score).

Source of Data: Platform genetic constraint widget (gnomAD)

Scoring:

A score from -1 to 1 is given to genes depending on their LOEUF metric rank, being -1 the least tolerant to LoF variation and 1 the most tolerant.

Mouse models

Definition: The international database Mouse Genome Informatics contains information about reported phenotypes when a gene is knocked-out in this animal model. These phenotypes are categorised in multiple phenotype classes, using an organ/system classification. We retrieve this information (available in our platform) and score the phenotypes classes regarding their severity (from 0 to -1). After aggregating all phenotypes with their scores according to the phenotype class they belong to, we use the harmonic sum to build a continuous score, which is normalised from 0 to -1.

Source of Data: Platform mouse phenotypes widget (Mouse Phenotypes, feeded from MGI, a reference database for mice knockouts)

Scoring:

• Below 0 to -1 = When the target has been knocked-out in mice there were multiple and severe phenotypes reported, with a score higher than the first quartile.

• 0 = Either the target has non-severe phenotypes reported or is in the first quartile of the normalised score.

• NA = No information available.

Gene essentiality

Definition: The second generation map of cancer dependencies (Pacini et al., 2024) increased the number of cancer cell lines analysed (930 CRISPR-Cas9 genome wide knock-out screenings, targeting almost 18,000 genes), spanning to 27 cancer types and curated patient genomic data, to identify cancer-type-specific and pan-cancer gene dependencies integrated with multi-omic markers.

Candidate anti-cancer therapeutic targets were characterised using a prioritisation criteria based on:

- Fitness Score. Strength of the effect on cellular fitness upon target depletion.

- Presence of dependency marker.

- Evidence linking the dependency and marker.

After applying a priority score based on approved drug targets, authors nominated 370 targets for 27 cancer types; 302 were cancer-type specific, while 196 where pan-cancer. This list of genes is the one used to label a target for gene essentiality.

Source of Data: A comprehensive clinically informed map of dependencies in cancer cells and framework for target prioritization. Pacini et al., 2024, Cancer Cell 42, 301–316. Supplementary Table 6. Gene essentiality widget (Cancer DepMap).

Scoring:

• -1 = Target reported as essential.

• 0 = Target not reported as essential

• NA = No information available

Known safety events

Definition: Target is associated with curated adverse events.

Source of Data: Safety liability data from Platform safety widget (Open Targets Safety) and Open Targets downstream analysis of toxicity datasets from PharmGKB.

Scoring:

• -1 = The target has at least one adverse event.

• NA = No information available.

Cancer driver gene

Definition: Target is classified as an oncogene and/or tumour suppressor gene.

Source of Data: Platform Cancer Hallmarks widget (COSMIC)

Scoring:

We use the attribute information from the cancer hallmarks, in the target profile. Here, targets considered as "cancer driver genes" are flagged as tumour suppressor, oncogene, or both

• -1 = Target is catalogued as driver gene (tumour suppressor, oncogene or both).

• NA = No information available.

Paralogues

Definition: Paralogue maximum identity percentage.

Source of Data: Platform comparative genomics widget (Ensembl Compara)

Scoring:

• Below 0 to -1 are linearly scored those targets with at least one paralogue in human harbouring at least 60% of identity with the target.

• 0 = Those targets with paralogues harbouring less than 60% of identity.

• NA = No information available about paralogues for that target.

Tissue specificity

Definition: HPA assessment on tissue-specific target expression.

Source of Data: Platform baseline expression widget (ExpressionAtlas, HPA and GTEx). We used the assessment for every target from the RNA expression data from the public version of Human Protein Atlas (proteinAtlasTissue)

Scoring:

Tissue distribution

Definition: HPA assessment on any detectable baseline expression for the target across tissues.

Source of Data: Platform baseline expression widget (Expression Atlas, HPA and GTEx). We used the assessment for every target from the RNA expression data from the public version of Human Protein Atlas.

Scoring:

To support more complex queries and advanced informatics workflows that use Google Cloud services, the Open Targets Platform data is also available as a Google Cloud public dataset via our Google BigQuery instance — open-targets-prod.

What is Google BigQuery?

Google BigQuery is a data warehouse that enables researchers to run super-fast, asynchronous SQL queries using Google's cloud infrastructure. After running your query, you can either export into various formats or copy into a Google Cloud bucket for further downstream analyses.

Open Targets Platform data is publicly accessible as a Google Cloud public dataset. Users only pay for the queries they perform on the data, and through this program, the first 1 TB per month is free.

BigQuery access points

Open Targets has uploaded all of our data to Google BigQuery. You can run queries via:

• Cloud Console

• Command line bq tool

• Client libraries, including Python

For more information on BiqQuery, please review the BigQuery documentation.

Example BigQuery SQL queries

Below is a sample query that uses our association_overall_direct dataset to return a list of targets associated with psoriasis (EFO_0000676) and the overall association score.

SELECT
  associations.targetId AS target_id,
  targets.approvedSymbol AS target_approved_symbol,
  associations.diseaseId AS disease_id,
  diseases.name AS disease_name,
  associations.score AS overall_association_score
FROM
  `open-targets-prod.platform.association_overall_direct` AS associations
JOIN
  `open-targets-prod.platform.disease` AS diseases
ON
  associations.diseaseId = diseases.id
JOIN
  `open-targets-prod.platform.target` AS targets
ON
  associations.targetId = targets.id
WHERE
  associations.diseaseId='EFO_0000676'
ORDER BY
  associations.score DESC

Similarly, you can use our drug_molecule dataset and pass a list of drug trade names to find relevant information:

DECLARE
  my_drug_list ARRAY<STRING>;
SET
  my_drug_list = [ 'Premarin',
  'Calcium disodium versenate',
  'Keytruda',
  'Vioxx',
  'Humira' ];
SELECT
  id AS drug_id,
  name AS drug_chembl_name,
  tradeNameList.element AS drug_trade_name,
  drugType AS drug_type,
  isApproved AS drug_is_approved,
  blackBoxWarning AS drug_blackbox_warning,
  hasBeenWithdrawn AS drug_withdrawn,
FROM
  `open-targets-prod.platform.drug_molecule`,
  UNNEST (tradeNames.list) AS tradeNameList
WHERE
  (tradeNameList.element) IN UNNEST(my_drug_list)

Tutorials and how-to guides

For more information on how to use BigQuery to access Platform data and example queries based on actual use cases and research questions, check out the Open Targets Community and our Google Cloud dataset homepage.

Do you have questions for us? Ask on Open Targets Community!

• Why can't I find a variant that I search for?

In the Open Targets Platform, a variant refers to any human variation that is associated with a disease, trait or phenotype that has been reported in at least one of our variant-to-phenotype sources. This accounts for only 1% of the total variants reported in gnomAD (6.5M vs 700M). See here for more information.

• Why can't I find a study I search for?

Each study undergoes quality control and validation procedures before ingestion into the Platform.

We remove studies that had unsupported study types, invalid target ID or invalid biosample ID (for molQTLs), and invalid ontology trait mapping. See here for study inclusion criteria.

For GWAS Catalog studies with summary statistic, we perform summary statistic quality control. If this fails, we exclude it.

For GWAS Catalog curated associations (top hits), we tightened the p-value significance threshold compared to OTG 22.10 (1e-8 instead of 5e-8). This may also lead to differences in the study list.

• What is a Credible Set?

A credible set is the set of variants near a genetic association signal that have a 95% probability of containing the true causal variant(s) for that signal.

• What are the quality control criteria for fine mapping?

Credible sets (CSs) are filtered out based on the following criteria:

1. The lead variant maps within the MHC region (chr6:25726063-33400556)

2. The lead variant has invalid chromosome coding (1:22, X, Y, XY, mt)

3. The CS is not on the list of valid studies

4. It is the GWAS Catalog Summary Statistics PICS CSs and it has valid SuSiE CS from the same region and study

5. It is the GWAS Catalog Curated Association PICS CS and it has valid GWAS Catalog Summary Statistics PICS CS from the same region and study

6. The sum of PIPs in the CS is not within the [0.95, 1] range

7. The CS didn't pass the study-specific p-value threshold (e.g. GWAS Catalog Curated Association p-value<=1e-8)

• Why is the V2G score not available anymore?

The variant-to-gene (V2G) score was developed as an aggregated score for the assignment of the variant to the gene in the region using different sources, such as association with molQTLs and VEP.

Although intuitive, V2G does not reflect the complexity of modern data.

We have therefore introduced a new score - Locus-to-Gene score (L2G), based on a machine learning algorithm that assesses the assignment score of the credible set (not only variant) to gene and does not use the old V2G pipeline anymore. It is a more accurate and sophisticated way of assigning genes to associated GWAS loci. In some cases, the L2G could be interpreted as V2G if the credible set is the size of only one variant. In the new Platform datasets, the L2G score has completely superseded V2G.

Full information about variant assignment to genes based on distance or VEP is available on all variant pages in the corresponding widgets. There is also information about whether the variant is part of the molQTL credible set, which can also be used to assign this variant to the gene of interest (available in the molQTL widget).

• Why are some L2G associations I found in OT Genetics 22.10 not found in the L2G predictions in OT Platform 25.03?

There could be several reasons for this:

1. We don't have the matching study anymore (due to quality control or validation).

2. We don't have the corresponding credible set anymore due to a different fine-mapping approach or p-value filtering thresholds (e.g. for GWAS Catalog curated associations we now use p-value < 1e-8 instead of p-value < 5e-8 in OTG 22.10).

3. The L2G model is different and can lead to different L2G estimates even when the same study and variant are presented. We don't display L2G assignments when L2G < 0.05.

• Where can I find studies or credible sets excluded from the Platform?

To ensure high-quality outputs, the data processing pipelines perform a number of validation steps across different datasets. For data provenance considerations the excluded part of the datasets are also available both on FTP (ftp://ftp.ebi.ac.uk/pub/databases/opentargets/platform/{release}/excluded) and Google Cloud (gs://open-targets-data-releases/{release}/excluded/). In this folder you'll find the list of excluded credible sets (credible_set), evidence (evidence), interactions (interaction), studies (study) and target validation (target_validation) dataset. These datasets also provide context about the reason for exclusion.

Based on genetic and functional genomics traits, the Locus-to-Gene (L2G) machine-learning algorithm ranks the most likely causative genes at each GWAS credible set. The likelihood that a gene is causal for a particular GWAS locus is measured by the L2G score, ranging from 0 to 1.

Features

📖 Gentropy

The predictive features used by the L2G algorithm are designed to capture various genetic and genomic contexts that influence the likelihood of a gene being causal at a given GWAS credible set.

Features are divided into four categories: