Computational Strategy Links Gene to Congenital Heart Problems

Researchers at the Gladstone Institutes in California have developed a new strategy for sifting through large-scale genetic data to better understand gene variant combinations that underlie congenital heart problems.  The approach combines techniques from genetics, computational biology, stem cell biology, and proteomics.

“A better understanding of the genetic basis of congenital heart disease could point to new strategies for not only blocking the development of the disease, which is currently very challenging, but also for alleviating issues that persist after surgery in order to improve quality and length of life,” says Bárbara González Terán, PhD, lead author of the study and a postdoctoral scholar at Gladstone.

Nearly 1% of all children are born with congenital heart disease. Abnormal versions of genes involved in the formation of the heart in the womb are believed to be the cause, but much remains to be learned about exactly which genes contribute to congenital heart disease and how they interact with each other.

The researchers say the power of their new method lies in its ability to provide insights into how combinations of variants—rather than single variants on their own—work together to cause congenital heart disease. “Previous methods have generated long lists of variants detected in patients, but many actually turned out to be inconsequential, so a major challenge in the field has been identifying which variants are most important,” says Gladstone President and Senior Investigator Deepak Srivastava, MD.  “Our approach pinpoints variants that are most likely to be involved in disease, allowing us to focus on those variants, deepen understanding of the underlying biology of the disease, and, we hope, move more rapidly toward new treatments.”

To identify potential culprit genes, the researchers carefully mapped out the entire network of interactions between the GATA4 and TBX5 proteins—already known to be required for healthy human heart formation—using precursor heart cells grown from human induced pluripotent stem cells. Next, they cross-referenced this 273-protein network with DNA sequencing data from over 3,000 children with congenital heart disease and their parents, developed by a NIH–funded consortium. Several dozen variants in the children’s sequencing data matched specific proteins also found in the GATA4-TBX5 network, singling them out as candidates that may contribute to congenital heart disease.

Determining whether each of the candidate variants identified in the GATA4-TBX5 network actually contribute to heart disease would involve years of research, but Maureen Pittman, a graduate student working on the study, developed a computational tool that ranked the candidates according to their likelihood of contributing to congenital heart disease. This ranking algorithm takes into account characteristics of the variant, the affected gene, and the type of heart defect found in patients with the variant. “Of the top-ranking variants we identified with the algorithm, some were in genes already known to contribute to congenital heart defects,” says Pittman. “But many had never before been linked to heart development, including a protein called GLYR1, which is involved in turning other genes on and off.”

Additional experiments in cells and mice indicated that GLYR1 indeed plays a central role in the formation of the heart in the womb, and a patient variant of GLYR1 disrupted heart development by hampering its interaction with GATA4.

While identifying GLYR1 as a key gene in heart development opens up a whole new biological space for understanding congenital heart disease, there is more work to be done. “Rarely is this disease caused by a single gene; a patient with the GLYR1 variant, for instance, could perhaps have additional variants inherited from their parents that by themselves were not enough to cause disease, but do so alongside the GLYR1 variant,” says Katie Pollard, PhD, director of the Gladstone Institute of Data Science and Biotechnology. “Our new approach could help identify specific combinations of variants that cause heart defects.”

The novel method, described in Cell, could also be adapted to identify combinations of variants that may underlie other complex diseases.  

“With more and more sequencing data being generated every year from patients with complex diseases, our approach will help guide where to focus among all the detected variants,” Srivastava adds.