Genomic Analysis: Overcoming a Formidable Challenge

Despite the improvements in scale and speed of generating genomic sequencing data, the challenge of genomic analysis and its costs remains.

“In my opinion, the cost of data analysis has always been the largest component of the overall cost,” says Elaine Mardis, PhD, co-executive director of the Steve and Cindy Rasmussen Institute for Genomic Medicine and Rasmussen Nationwide Foundation Endowed Chair in Genomic Medicine at Nationwide Children’s Hospital. She also is a professor of Pediatrics at The Ohio State University College of Medicine. “And this has only changed marginally since the advent of next generation sequencing.”

Some of the many steps to genomic analysis have gotten faster, but the painstaking process of looking across the approximately 6 billion data points generated by each genome sequence for alterations remains a lengthy and time-consuming process that requires special training and expertise.

“This is especially true when we are clinically interpreting the whole genome data and using that information to diagnose and identify best care for patients,” says Dr. Mardis.

Varhouse: A Technological Piece of the Solution

Further streamlining and accelerating analytical workflows within IGM requires both automation and collaboration. The Office of Data Sciences, led by Peter White, PhD, chief data sciences officer at Nationwide Children’s, plays a critical role in addressing genomic analysis challenges. Dr. White has been at the forefront of solving these complex problems.

Now, Dr. White and his team of software engineers have developed the Variant Analysis Warehouse — known as Varhouse — a system designed to store, manage, and interpret genomic data.

“We’ve solved the challenges of processing a patient’s DNA sample through genome sequencing and identifying all the genetic variants within that genome,” explains Dr. White. “Varhouse addresses the final piece of the puzzle — determining which of the millions of genetics variants are actually responsible for a patient’s disease and relevant to their treatment.”

Varhouse offers cloud-enabled, scalable storage and real-time genomic analysis. It enables identification of genomic alterations that are linked to the patient’s diagnosis, enabling the delineation of therapeutic approaches or other aspects of care.

Variant Experts: A Human Piece of the Solution

Another asset that is enabling the Institute for Genomic Medicine to improve analysis speeds while maintaining high fidelity, accurate reporting is the Clinical Genomic Analysis Team (GCAT) composed of variant scientists and analysts who are the primary reviewers of genomic sequencing data. This group of experts is involved in identifying disease-related alterations from genomic data, interpreting these data, and drafting the clinical reports that describe these alterations. They are also integral to validation of new software tools for genomic analysis.

“We are incredibly fortunate to have a team of highly skilled variant scientists and analysts within our clinical laboratory,” says Catherine Cottrell, PhD, FACMG, section chief of the Institute for Genomic Medicine Clinical Laboratory at Nationwide Children’s. “Other clinical directors have come to us asking for advice in how to get institutional support for similar teams. Variant analysis is a growing and often underrepresented area in genomics. However, we are seeing greater recognition of the expertise variant scientists can bring. Our lead variant scientist, Vijay Jayaraman, is featured as an invited speaker at the 2025 American College of Medical Genetics Annual Meeting on this very topic.”

Three days is the fastest clinical turn-around for rapid genome sequencing at Institute for Genomic Medicine (from time of sample receipt to result). This includes time for test accessioning into the lab information system, DNA extraction from the blood sample, documenting clinical features of the patient, preparing the sample for sequencing (termed library preparation), quality control checks throughout the process, the sequencing data generation, a multi-step bioinformatic analysis, variant interpretation through Varhouse and finally preliminary reporting of the clinical result.

“I believe the biggest gain in the future is more rapid bioinformatic analysis. This would enable annotated data to be in the hands of our variant analysts and scientists sooner, speeding up the interpretation and reporting by a clinical director,” adds Dr. Cottrell, who is also a clinical professor within the Departments of Pathology and Pediatrics at The Ohio State University.

Integration for Success

The integrated nature of the IGM is another strength that supports advanced analysis, says Beth Kozel, MD, PhD, chief of the Division of Genetics and Genomic Medicine and the director of Constitutional Genomics Translational Research in the Institute for Genomic Medicine.

“To me, this is where one of the strengths of our program shines. The lab team can come up with a list of high priority variants and actually talk to the clinical team about the fit when there is gray area,” she says. “And if the fit is good, we can sometimes do other testing types to move us from a “maybe” to something we are more confident about. That is truly at the intersection of clinical care and research. That intersection is exactly what we are going to need for the next decade to clean up the ‘noise’ of conflicting information in the system. The informatics piece will help us do some of that work faster, but the devil really is in the details for these things and so it is important to balance the bulk data approaches with the painstaking n of 1 overview to optimize our success.”

Reanalysis: Solving Unsolved Cases

While 28% of Nationwide Children’s rGS tests returned actionable results, that means the majority still didn’t turn up conclusive answers for the family. The same goes for other genomic and genetic tests that don’t offer the answers hoped for. These tests may have ruled out some diagnoses, but the case as a whole may remain unsolved.

To address this challenge of rapidly advancing new information, the Institute for Genomic Medicine currently offers Genome Sequencing Reanalysis for patients who previously had clinical genome sequencing or clinical rGS performed by Nationwide Children’s Institute for Genomic Medicine Clinical Laboratory. As technology and analysis capacity continue to scale, the team aims to streamline reanalysis processes so that new information in the literature and the databases can get to clinicians as quickly as possible.

“Unsolved patient cases, especially when you strongly suspect a genetic cause, can be highly vexing,” says Dr. Cottrell. “We work in partnership with our clinical colleagues to reanalyze existing genomic data for these patients. Knowledge of gene and disease associations grows every day and enabling the opportunity to find an answer.

However, reanalysis currently depends on a provider knowing when to request a review from the IGM — a process that can be challenging given the pace of new discoveries. Dr. White and his team in the Office of Data Sciences are now exploring how artificial intelligence (AI) could change the game for reanalysis.

A new collaborative initiative called GENIUS — short for Genomic Evaluation using Next-generation Intelligence for Understanding & Swift Diagnosis — brings together experts from Institute for Genomic Medicine, Clinical Genetics and Genomics, and the Office of Data Sciences to tackle a growing genomic analysis challenge: what to do for patients who remain undiagnosed after initial testing.

“About 8,000 rare diseases have been described, and every month, an additional 40 to 50 new rare disease genes are discovered,” explains Dr. White. “It’s not reasonable to expect our clinicians to stay on top of that volume of new literature. So, our team is evaluating how AI can help identify patients who may now have a diagnosable condition.”

GENIUS is a cutting-edge genomic medicine platform designed to enhance diagnosis and transform neonatal care. It combines machine learning (ML) and natural language processing (NLP) to identify high-risk patients and automate the reanalysis of genomic data. By scanning electronic health records in real time, GENIUS can alert providers when rapid genomic sequencing may be warranted.

As part of its reanalysis capability, GENIUS leverages Varhouse’s capability to continuously retrieve the latest public and clinical genomic annotations — ensuring that every genome is interpreted with the most current information available. Together, GENIUS and Varhouse exemplify how intelligent systems can help clinicians deliver faster, more accurate diagnoses.