An integrated clinical program and crowdsourcing strategy for genomic sequencing and Mendelian disease gene discovery

Abstract

Despite major progress in defining the genetic basis of Mendelian disorders, the molecular etiology of many cases remains unknown. Patients with these undiagnosed disorders often have complex presentations and require treatment by multiple health care specialists. Here, we describe an integrated clinical diagnostic and research program using whole-exome and whole-genome sequencing (WES/WGS) for Mendelian disease gene discovery. This program employs specific case ascertainment parameters, a WES/WGS computational analysis pipeline that is optimized for Mendelian disease gene discovery with variant callers tuned to specific inheritance modes, an interdisciplinary crowdsourcing strategy for genomic sequence analysis, matchmaking for additional cases, and integration of the findings regarding gene causality with the clinical management plan. The interdisciplinary gene discovery team includes clinical, computational, and experimental biomedical specialists who interact to identify the genetic etiology of the disease, and when so warranted, to devise improved or novel treatments for affected patients. This program effectively integrates the clinical and research missions of an academic medical center and affords both diagnostic and therapeutic options for patients suffering from genetic disease. It may therefore be germane to other academic medical institutions engaged in implementing genomic medicine programs.

Fig. 1

Workflow overview. The workflow begins with the clinical assessment of cases, where referring physicians present cases of presumptive unknown monogenic etiology from their clinical practice for the collective development of a case solution strategy. Based on the inferred inheritance mode, the most informative family members are selected for genomic sequencing followed by analysis of the WES/WGS data using a computational pipeline designed to identify rare Mendelian variants. A final candidate gene list is prioritized using in-house bioinformatic tools, literature surveys, and crowdsourcing. The final candidate gene is confirmed by segregation analysis, matchmaking for second case hits, and by in vitro and in vivo functional studies. Both genomics and functional biology thus inform the diagnosis and clinical management of individual patients, while unique patient conditions provide insight into gene and pathway function

Fig. 2

Overview of the analytical pipeline. a The computational pipeline is shown, including discrete steps for the alignment of read data, joint genotyping, variant prioritization programs, incorporation of external data sources, and variant visualization, culminating in downstream analyses; b detailed breakdown of the upstream analysis is shown, including the production of validated, annotated, and prioritized candidate variants; c detailed breakdown of the downstream analysis culminates in the evaluation of candidate genes and the variants within them for causality for the condition of interest

Fig. 3

Crowdsourcing of Mendelian cases. Clinical and genomic data of cases under analysis are presented in fully de-identified format via a secure portal and in compliance with HIPAA and patient privacy regulations. This crowdsourcing mechanism provides clinicians, researchers, and data analysts with the opportunity to “interactively” analyze the data, vet analytical approaches, explore follow-up options, and to obtain second, third, or fourth opinions before finalizing a particular analytical and validation strategy. In some cases, the crowdsourcing strategy can serve as a matchmaker for a second case based on the phenotype, genotype, or both