Benjamin Solomon, M.D.

Genetic conditions, though often individually rare, are common in aggregate, and disproportionately impact patients, families, healthcare systems, and society. New and evolving technological approaches continue to yield dramatic changes in the ability to diagnose (and therefore better manage) affected individuals. The capacity to efficiently analyze - as well as produce - genomic data has also increased significantly. Substantial work remains to enable optimal mining of relevant phenotypic as well as genotypic data.

Recent developments in applying artificial intelligence (AI) and related computational approaches – including deep learning (DL) - to health-related questions suggest strong potential. While these approaches have specifically been applied to genetic conditions, many challenges and questions require further investigation to optimally utilize these methods.

The overall goal of the Medical Genomics Unit (MGU) is to use AI and DL-based analytic and computational approaches to develop, test, and implement methods of interrogating datasets relevant to genetic conditions. Types of data to be analyzed include clinical/phenotypic information such as images, data from medical records/electronic health records, as well as genomic and related biologic data. Combining methods to incorporate multiple types of data is anticipated to enable novel insights as well as flexible models of analysis. Improving these methods may allow more efficient and accurate investigations into aspects of many aspects of human health and disease.

These methods and the related results will focus on exploring questions that are directly clinically-relevant in addition to scientifically important, with the goal of yielding or helping lead to tools that can assist researchers to understand genetic conditions and clinicians to better manage patients and families. While some MGU investigations will be conducted using data from specific conditions in proof-of-principle scenarios, the general methods are anticipated to be initially largely agnostic to any particular condition. That is, these methods will enable approaches that can be applied broadly to many types of appropriate datasets.

Dr. Solomon earned his medical degree from Geisel School of Medicine at Dartmouth. He completed medical training in pediatrics and clinical genetics through a joint Children's National/National Human Genome Research Institute program. After completing this program, Dr. Solomon remained at NHGRI for several years as a Staff Clinician; his work focused on understanding the causes and biology of a number of congenital disorders as well as applying emerging technologies and analytic approaches to genomic and phenotypic datasets in order to ask a variety of research questions.

Dr. Solomon joined the Inova Translational Medicine Institute (ITMI) in 2013 as the Chief of the Division of Medical Genomics. In this role, Dr. Solomon led a team of clinicians, bioinformaticists, and bench-based scientists to deliver clinical care and conduct genomic research. In 2016, Dr. Solomon became the Managing Director of GeneDx, a genetics/genomics diagnostic company with a strong emphasis on research and the discovery of novel causes of disease. In this role, which he held until returning to NHGRI as Clinical Director, he led a team of over 400 molecular geneticists, genetic counselors, and laboratory and research staff. Throughout his entire career, he has always continued to be an avid participant in clinical genomics education, including formal and informal teaching of medical students, residents, fellows, postdoctoral and other trainees, as well as re-educating the existing clinical and scientific workforce about genomics. Additionally, since 2017, Dr. Solomon served as deputy editor-in-chief and then editor-in-chief for the American Journal of Medical Genetics. He has authored and co-authored more than 150 peer-reviewed papers, commentaries, and book chapters. His research interests focus on developing and testing methods to accurately and efficiently analyze diverse datasets relevant to the causes, manifestations, and management of genetic conditions, and which can be applied to broader areas of human health and disease.