Developments in Genome Analysis

Abstract:
The evolution of new sequencing technologies has led to the generation of massive volumes of data which has outpaced our ability to analyze, interpret and make use of, to inform patient management. I will discuss how my laboratory analyzes and reports on genomic results with examples from my clinical laboratory and the GENCOV/HostSeq study. In addition, I’ll discuss challenges we face in sharing data generated from the clinical laboratory; new and upcoming projects that aim to develop a Canadian Genomics Health Data Ecosystem; the generation of new genomic databases and novel computing platforms that will help inform clinical decision making in molecular diagnostics and to fuel biomedical research for years to come.

.

Abstract:
Although computational and experimental advances have enabled large-scale determination of sequence variant effects, pathogenic variants may have incomplete penetrance or variable expressivity, with effects that depend on environmental or genetic context. Cell-based (or other in vitro) assays offer an opportunity to model these contexts. The diversity of human phenotypes that can be caused by different pathogenic variants in the same gene can also be modeled, not only based on quantitative results from a given assay, but also through the use of multiple qualitatively different assays for the same gene. For multiple human proteins, e.g., MTHFR and LDLR, I will give examples that represent the emerging computational/experimental hybrid field of contextual variant effect mapping.

Abstract:
Genetic variants in untranslated regions (UTRs) contribute to rare disease, but interpreting their functional impacts poses a major challenge. The utility of predictive models for UTR variants’ pathogenicity classification remains unclear. We will discuss the assessment of deep learning models using a curated set of pathogenic and likely pathogenic (P/LP) variants in the 3’ and 5’ UTRs. Key considerations in UTR variant classification and DL models’ mechanistic insights will be highlighted. We will also introduce Deep Genomics’ RNA foundation model, BigRNA, which is adept at discerning pathogenic effects associated with UTR variants across diverse mechanisms. BigRNA holds immense potential for advancing the prediction of non-coding variant effects in the scientific community. 

Despite continued advances in sequencing, clinical genetic testing currently relies on multiple testing modalities to access known genomic variation, complicating integrated clinical reporting, and often necessitating the utilization of multiple reference laboratories. Leveraging HiFi long read genome sequencing (lrGS), a single test can evaluate methylation, copy number variation, structural variation, expansion disorders, X-activation studies, telomere length, and single nucleotide variation with haplotype phasing. To evaluate the impact of a single assay we assess the first 1500 HiFi genomes in a rare disease cohort (“Genomic Answers for Kids” / GA4K). Analyses included copy number, structural variation, single nucleotide variation, repeat expansion, and for a subset of genomes 5-methyl C detection and X-inactivation. Each layer of the lrGS calling pipeline has been demonstrated to reveal new, previously undiagnosed pathogenic variation. Altogether, the implementation of lrGS in an ES/GS negative cohort results in up to a 10% increase in diagnostic yield. Importantly, known variants were recapitulated by machine-learning based interpretation, indicating lrGS could be utilized as a first-tier genome test, simplifying genetic testing algorithms. In addition, novel capabilities of lrGS augment rare disease analyses beyond increasing current diagnostic rate: discovery of non-coding rare variation leading to hypermethylation, personal assemblies augmenting rare structural variant calling and high-resolution analyses of instability of STRs in candidate disease loci.

Abstract:
There is broad concern that genomic technologies may not reach those with the greatest health disparities. There are numerous reasons why there are barriers to specialty care including genetic/genomic diagnosis for Indigenous people of Canada. Of further concern is the lack of Indigenous reference data in public databases limiting, interpretation of genetic/genomic sequencing results. As will be shown in this presentation, Indigenous involvement and on-going Indigenous governance of data has the potential to change the trajectory, improving access to diagnosis for Indigenous patients with genetic conditions. The current status of the IBVL, the plans for implementation, and sustainability will also be discussed.

Abstract:
In this seminar I will discuss the current applications for RNA sequencing as a diagnostic tool in mendelian disease and highlight emerging research in the field.

Abstract:
Our clinic had a goal to provide Canadians with an option for high-quality, comprehensive clinically responsible whole genome sequencing designed for proactive screening purposes. With no commercially available laboratory options meeting our criteria for this program, we built our own in collaboration with the Hospital for Sick Children. This program launched in February 2022. Eighteen months later, we have had nearly 100 Whole Genome Sequencing (WGS) results reported. Data collection is ongoing; here we discuss genetic counselling considerations with this program and we report on our experience and results to date.

More info coming soon