Facilitating collaboration in rare genetic disorders through effective matchmaking in DECIPHER

Abstract

DECIPHER (https://decipher.sanger.ac.uk) is a web-based platform for secure deposition, analysis, and sharing of plausibly pathogenic genomic variants from well-phenotyped patients suffering from genetic disorders. DECIPHER aids clinical interpretation of these rare sequence and copy-number variants by providing tools for variant analysis and identification of other patients exhibiting similar genotype-phenotype characteristics. DECIPHER also provides mechanisms to encourage collaboration among a global community of clinical centers and researchers, as well as exchange of information between clinicians and researchers within a consortium, to accelerate discovery and diagnosis. DECIPHER has contributed to matchmaking efforts by enabling the global clinical genetics community to identify many previously undiagnosed syndromes and new disease genes, and has facilitated the publication of over 700 peer-reviewed scientific publications since 2004. At the time of writing, DECIPHER contains anonymized data from ∼250 registered centers on more than 51,500 patients (∼18000 patients with consent for data sharing and ∼25000 anonymized records shared privately). In this paper, we describe salient features of the platform, with special emphasis on the tools and processes that aid interpretation, sharing, and effective matchmaking with other data held in the database and that make DECIPHER an invaluable clinical and research resource.

Keywords: MatchMaker Exchange; genetic disorders; genotype-phenotype correlation; rare diseases.

Figure 1

A: The DECIPHER community consists of centers of clinical genetics and associated research groups. Each center has a designated clinical coordinator and authorized users with distinct roles and access permissions that allow them to access, edit, and/or deposit data. B: The DECIPHER database contains over 17,000 anonymized patient‐consented records (genotype and phenotype) available to all users for matchmaking, contact, or collaboration. C: A metric of the success of collaboration is the increased number of peer‐reviewed publications that have used data from DECIPHER for the identification of previously unidentified syndromes and/or novel genes implicated in genetic disorders.

Figure 2

DECIPHER data entry and analysis interface. A: Phenotype entry in DECIPHER is based on the Human Phenotype Ontology (HPO) tree and provides a simple drag‐and‐drop interface to quickly record observed or absent patient and parental phenotypes. B: DECIPHER search allows querying the database using a combination of terms and is aided by an autosuggest facility to quickly find term of interest. C: An interactive histogram representation of the karyotype for a search on phenotype “seizures” showing a peak in the region of 2q24.3 encompassing the SCN2A gene. Individual inverted triangles represent sequence variants, whereas copy‐number variants are shown in red (deletion) and blue (duplications). D: An ARID1B gene sequence variant in DECIPHER and other records in the database overlapping the same gene. DECIPHER assists the identification of other patients in the database that overlap a position using a functional overlap score and highlights shared phenotypes in these patients. Pie charts provide a visual aid to the breakdown of the positional overlap based on various criteria.

Figure 3

Facilitating matchmaking in DECIPHER. A: Registered depositors of the DECIPHER platform can deposit their patient data into their center and then find other anonymized patient‐consented data similar to their patient in the database and make direct contact with other depositors. Anonymous patient‐consented data are also shared with Ensembl and UCSC genome browsers. External users can find patient‐consented data via search or Ensembl and UCSC genome browsers and initiate contact with a depositor for information or collaboration facilitated by DECIPHER. B: A mock‐up of the implementation of the MatchMaker Exchange (MME) application programming interface (API) in DECIPHER. Depositors who share their anonymous data will benefit from finding other patients in other databases that are part of the MME API project.