Centers for Mendelian Genomics: A decade of facilitating gene discovery

Samantha Baxter, Jennifer E. Posey, Nicole J. Lake, Nara Sobreira, Jessica X. Chong, Steven Buyske, Elizabeth Blue, Lisa H. Chadwick, Zeynep Coban‐Akdemir, Kimberly F. Doheny, Colleen Davis, Monkol Lek, Matthew T. Wheeler, Shalini N. Jhangiani, Mark Gerstein, Richard A. Gibbs, Richard P. Lifton, Daniel G. MacArthur, Tara C. Matise, James R. Lupski, David Valle, Michael J. Bamshad, Ada Hamosh, Shrikant Mane, Deborah A. Nickerson, Heidi L. Rehm, Anne O’Donnell‐LuriamedRxiv 2021 · DOI

Abstract

Mendelian disease genomic research has undergone a massive transformation over the last decade. With increasing availability of exome and genome sequencing, the role of Mendelian research has expanded beyond data collection, sequencing, and analysis to worldwide data sharing and collaboration. Over the last 10 years, the NIH-supported Centers for Mendelian Genomics (CMGs) have played a major role in this research and clinical evolution. We highlight the cumulative gene discoveries facilitated by the program, biomedical research leveraged by the approach, and the larger impact on the research community. Mendelian genomic research extends beyond generating lists of gene-phenotype relationships, it includes developing tools, training the larger community to use these tools and approaches, and facilitating collaboration through data sharing. Thus, the CMGs have also focused on creating resources, tools, and training for the larger community to foster the understanding of genes and genome variation. The CMGs have participated in a wide range of data sharing activities, including deposition of all eligible CMG data into AnVIL (NHGRI’s Genomic Data Science Analysis, Visualization, and Informatics Lab-Space), sharing candidate genes through Matchmaker Exchange (MME) and the CMG website, and sharing variants in Geno2MP and VariantMatcher. The research genomics output remains exploratory with evidence that thousands of disease genes, in which variant alleles contribute to disease, remain undiscovered, and many patients with rare disease remain molecularly undiagnosed. Strengthening communication between research and clinical labs, continued development and sharing of knowledge and tools required for solving previously unsolved cases, and improving access to data sets, including high-quality metadata, are all required to continue to advance Mendelian genomics research and continue to leverage the Human Genome Project for basic biomedical science research and clinical utility.

Introduction

While individual Mendelian (single-gene) diseases are rare, they collectively affect millions of people worldwide. Historically, identifying the genetic cause of these conditions was a slow, agonizing process involving positional cloning and physical mapping. The advent of Next-Generation Sequencing (NGS) in the early 2010s promised a revolution, but individual researchers often lacked the "statistical power"—specifically, enough unrelated patients with the same symptoms and mutations—to definitively link a candidate gene to a disease.

The Centers for Mendelian Genomics (CMG) program was established to bridge this gap. It addressed the challenge of how to coordinate global expertise, standardize massive genomic datasets, and facilitate the "matching" of rare cases across international borders to accelerate the discovery of the thousands of disease-causing genes that remain hidden in the human genome. Over the last decade, the CMG has transformed rare disease research from isolated lab efforts into a global, data-sharing powerhouse.

Methods and Data Sharing Infrastructure

The CMG approach shifted the paradigm from "siloed research" to "distributive research." The centers provided high-throughput Exome (ES) and Genome (GS) sequencing and rigorous bioinformatics pipelines, while a global network of collaborators provided deep clinical phenotyping. Central to this success was the formalization of three distinct modes of data sharing:

  1. Two-sided matching: Connecting two researchers who both have a candidate gene (e.g., via Matchmaker Exchange).
  2. One-sided matching: Allowing a researcher with a candidate gene to query a database of primary genomic data (e.g., Geno2MP).
  3. Zero-sided matching: Providing credentialed researchers access to raw, read-level data (CRAM files) to perform cohort-wide discovery without prior candidates (e.g., via the AnVIL cloud platform).

Furthermore, the program developed open-source tools like PhenoDB and seqr, which facilitate real-time, collaborative variant filtration and analysis across the research community. The program also heavily utilized and contributed to community resources like gnomAD and ClinVar.

Results and Cumulative Impact

The CMG program has managed a massive decade-long data collection effort, resulting in a cohort of 28,991 families. This includes 75,573 exomes, 3,876 genomes, 714 transcriptomes (RNA-seq), and 385 methylation arrays. Over 60 terabytes of this data now reside in the AnVIL platform for community access.

Key Discovery Metrics

  • Discovery Volume: The CMG identified 1,637 new gene-disease relationships (Tier 1) and 2,270 strong candidate genes (Tier 2).
  • Global Reach: The program collaborated with 11,771 researchers across 2,773 institutions in 90 countries.
  • Clinical Impact: Over 11,000 families received a confirmed or potential genetic answer. Remarkably, 65% of the CMG’s novel gene discoveries already have "Pathogenic" or "Likely Pathogenic" variants submitted to ClinVar by clinical labs, showing rapid integration into medical diagnostics.

Scientific Insights

Analysis of this decade of data reveals that collaboration is the primary engine of discovery; over 90% of CMG publications involved more than one institution. The program also highlighted the "long tail" of phenotypes, identifying 778 phenotypic expansions where known disease genes were found to cause previously unrecorded symptoms. Discovery rates were notably highest in neurodevelopmental, skeletal, and immune system disorders. Additionally, periodic re-analysis of "unsolved" cases using updated algorithms and gene lists yielded new diagnoses in up to one-third of cases.

Discussion and Future Directions

Despite these significant achievements, several hurdles remain. A major limitation is the diversity gap; 73% of collaborators and 67% of sequenced cases were of White/European descent. Expanding to underrepresented populations is a scientific and ethical necessity to ensure the benefits of genomic medicine are universal.

Furthermore, over half of the families sequenced by the CMG remain without a molecular diagnosis. These "cold cases" likely involve complex structural variants, non-coding mutations, or epigenetic changes that standard exome sequencing cannot detect. To address this, the program is transitioning into the Mendelian Genomics Research Consortium (MGRC). This next phase will focus on unsolved cases using long-read sequencing, functional "omics" (such as RNA-seq and proteomics), and advanced AI-driven pathogenicity predictions. Strengthening the communication between research and clinical labs remains essential to continue leveraging the Human Genome Project for basic science and clinical utility.

Summary updated

Abstract

Abstract Mendelian disease genomic research has undergone a massive transformation over the last decade. With increasing availability of exome and genome sequencing, the role of Mendelian research has expanded beyond data collection, sequencing, and analysis to worldwide data sharing and collaboration. Over the last 10 years, the NIH-supported Centers for Mendelian Genomics (CMGs) have played a major role in this research and clinical evolution. We highlight the cumulative gene discoveries facilitated by the program, biomedical research leveraged by the approach, and the larger impact on the research community. Mendelian genomic research extends beyond generating lists of gene-phenotype relationships, it includes developing tools, training the larger community to use these tools and approaches, and facilitating collaboration through data sharing. Thus, the CMGs have also focused on creating resources, tools, and training for the larger community to foster the understanding of genes and genome variation. The CMGs have participated in a wide range of data sharing activities, including deposition of all eligible CMG data into AnVIL (NHGRI’s Genomic Data Science Analysis, Visualization, and Informatics Lab-Space), sharing candidate genes through Matchmaker Exchange (MME) and the CMG website, and sharing variants in Geno2MP and VariantMatcher. The research genomics output remains exploratory with evidence that thousands of disease genes, in which variant alleles contribute to disease, remain undiscovered, and many patients with rare disease remain molecularly undiagnosed. Strengthening communication between research and clinical labs, continued development and sharing of knowledge and tools required for solving previously unsolved cases, and improving access to data sets, including high-quality metadata, are all required to continue to advance Mendelian genomics research and continue to leverage the Human Genome Project for basic biomedical science research and clinical utility.
BibTeX
@misc{baxter2021centers,
  author = {Samantha Baxter and Jennifer E. Posey and Nicole J. Lake and Nara Sobreira and Jessica X. Chong and Steven Buyske and Elizabeth Blue and Lisa H. Chadwick and Zeynep Coban‐Akdemir and Kimberly F. Doheny and Colleen Davis and Monkol Lek and Matthew T. Wheeler and Shalini N. Jhangiani and Mark Gerstein and Richard A. Gibbs and Richard P. Lifton and Daniel G. MacArthur and Tara C. Matise and James R. Lupski and David Valle and Michael J. Bamshad and Ada Hamosh and Shrikant Mane and Deborah A. Nickerson and Heidi L. Rehm and Anne O’Donnell‐Luria},
  title = {{Centers for Mendelian Genomics: A decade of facilitating gene discovery}},
  year = {2021},
  doi = {10.1101/2021.08.24.21261656},
  howpublished = {medRxiv},
}
Figure