Utilizing mammalian constraint to decipher human genetic disease and cancer

Evolutionary constraint is a unique identifier of functional sequences in the human genome. High constraint indicates genomic regions resistant to change over evolutionary time due to functionality.

The primary aim of this project is to utilise evolutionary constraint derived from 240 mammals to identify and prioritize variants that cause genetic disease and cancer.

Genome-wide association mapping has identified thousands of disease-associated regions, mainly regulatory, but it is still challenging to go beyond association and identify the disease-causing variants. Dogs are genetically homogenous and excellent spontaneous models for human disease.

We utilize this to map genes where the disease phenotypes show strong homology between species e.g., amyotrophic lateral sclerosis (ALS) and obsessive-compulsive disorder (OCD). Using evolutionary constraint, we can understand which variants are most likely to cause altered gene expression.

The same principle applies to somatic tumor mutations where non-coding mutations with high evolutionary constraint can point out novel cancer driver genes.

We will further use machine learning and perform large scale experimental analysis to evaluate candidate functional variants and the molecular mechanisms underlying disease.

This research will provide a unique opportunity to decipher the biology of human disease and, in the future, may help identify drugs that can be more effective in certain patients with mutations in specific genes.