Ryan is interested in the relationship between structure and function in the human genome. His research integrates genomics, physiology, and bioinformatics to address five principal questions:
Fig1a from Collins*, Brand*, et al., Nature, 2020
1. Structural Variation: How does genome content vary between individuals and cell types? The economization of whole-genome sequencing has produced a small sea of sequenced human genomes, and with it, the realization that nearly 1% of every human genome is structurally variant: while the exact nucleotide sequences between two individuals’ DNA are highly homologous (>99.9% identity), the large-scale organization of each genome is substantially more mutable, with thousands of segments between ~50bp and hundreds of kilobases being deleted, duplicated, inverted, or inserted in diverse arrangements. This concept, known as structural variation, is a technological challenge to capture and a taxonomic challenge to categorize. Carefully documenting these ubiquitous rearrangements in indivuals and across populations will be an essential step in generating a complete portrait of the human genome.
Fig4 from Collins*, Brand*, et al., Nature, 2020
2. Dosage Sensitivity: Why do some parts of the genome tolerate increased or decreased copies while others don’t? Natural selection maintains the vast majority of mammalian genomes as diploid (i.e., two copies of each chromosome). However, deletion and duplication of genomic segments—collectively known as copy-number variants (CNVs)—are well-described mechanisms for evolutionary adaptation, such as duplications of the human salivary amylase locus aiding the digestion of dietary starch. Yet examples of CNVs yielding clear adaptive advantages are relatively rare in humans. Instead, large CNVs have been widely associated with Mendelian and complex diseases, including famous examples like duplications of APP in early-onset familial Alzheimer’s disease and deletions of NRXN1 in schizophrenia. Likewise, in somatic tissues, large mosaic CNVs have been implicated in autism and abnormal clonal hematopoiesis, and somatic genome instability is a hallmark of many cancers. The influence of CNVs in human traits and diseases is therefore as broad as it is profound, but the principles determining why some loci tolerate CNVs while others don’t are mostly unknown.
Fig3b-d from Redin et al., Nat. Genet., 2017
3. Functional & Systems Genomics: What are the essential biochemical players and interactions in the human genome? Numerous studies have now cemented that regulation of gene expression is a convoluted and multifarious system refined by billions of years of evolution, and involves both the genome’s sequence and spatial topology as well as the transcriptome, epigenetic modifications, and essentially almost everything else present within a cell’s nucleus. While significant strides have landed towards understanding each of these layers in isolation, a remaining hurdle is the synthesis of these sub-systems into a comprehensive model of how genome architecture—specifically the structure and three-dimensional organization of DNA and chromatin—drives gene expression at the organismal, tissue, and cellular scales. Technological advances permitting high-resolution surveys of chromatin states, like Hi-C and 4C, have begun to illuminate the genome as a dynamic, three-dimensional, highly regulated biological system. These new data now allow us to ask how the composition of our chromosomes at the DNA level regulates and coordinates the remarkable process of compacting six linear feet of DNA, comprising over three billion nucleotides, into the nucleus of each of the trillions of cells in every human.
Fig5a-c from Collins et al., Genome Biol., 2017
4. Human Disease: Which changes in genome structure contribute to disease risk and pathogenesis? Central to Ryan’s research interests are human diseases or disorders with strong genetic components, such as congenital heart defects and neurodevelopmental disorders like autism. While understanding the dynamics of the genome’s architecture is an essential first step, the second—and perhaps more important—question to ask is how this architecture differs between affected and unaffected individuals. Elucidating trends in structural variation and genome regulation in human disease has the potential to expand our knowledge of the causes of many currently intractable human diseases and, with it, identify novel treatments and therapeutic targets.
GATK-SV prototype schematic (adapted from Collins*, Brand*, et al., Nature, 2020)
5. Bioinformatics & Biotechnologies: Can we develop computational methods and statistical models to learn more about genome structure and better predict, identify, and treat disease? Finally, a natural and necessary complement to investigating the human genome across its multiple biological dimensions is the advancement of computational methods and statistical frameworks to test previously untestable hypotheses. Particularly, with the slated sequencing of hundreds of thousands of human genomes over the next several years, efficient algorithms and statistical methods will be foundational in analyzing these data, and may reveal biological insights to ultimately guide clinical care.
Statement of Professional Objectives
Precision medicine will revolutionize the delivery of preventative and palliative healthcare in our lifetime. Deciphering the complex architecture of the human genome is the key to unlocking the precision medicine revolution, as the genome is the blueprint for each patient’s biochemical individuality. To this end, Ryan’s career goal is to expedite the arrival of routine precision medicine by advancing our understanding of the relationship between human genome structure and function.