NEW YORK – Using a deep learning approach to predict the effects of mutations on RNA-binding protein (RBP) target sites, researchers have found that RBP dysregulation is a principal contributor to individuals' risks of developing psychiatric disorders.
In a study published on Monday in Nature Genetics, researchers at Princeton University, Rockefeller University, and the Flatiron Institute noted that while there is a strong genetic basis for psychiatric disorders, the underlying molecular mechanisms are largely unmapped.
RBPs are known to be responsible for most post-transcriptional regulation, and act as key gatekeepers of cellular homeostasis, especially in the brain. RBP dysregulation, therefore, explains a significant amount of heritability for psychiatric disorders that is not captured by large-scale molecular quantitative trait loci studies, and has a stronger impact than common coding region variants, the researchers wrote.
The investigators were also able to use the genome-wide profiles of RBP dysregulation that were predicted by the deep learning approach to identify DDHD2 as a candidate schizophrenia risk gene.
They had previously evaluated the accuracy of the deep learning-based sequence model, Seqweaver, both computationally and experimentally, applying it to detect de novo noncoding mutation signals in autism in probands versus their unaffected siblings. In this study, they used Seqweaver to build a profile of allele-specific effects of inherited variants genome-wide in order to examine the diverse landscape and impact of RBP dysregulation in complex psychiatric disorders. They also made the annotations of variants linked to RBP dysregulation publicly available.
The researchers first sought to analyze selection traces that act on binding site sequences that would impact RBP regulation. They hypothesized that RBP target site variants that lead to dysregulation are also subject to negative selection, and tested this theory on human variants from control cohorts released by the Genome Aggregation Database. For each transcribed, noncoding variant in gnomAD, the researchers interrogated the levels of RBP dysregulation using deep learning inference based on 232 Seqweaver RBP models. They found significant depletion of strong-effect RBP dysregulation variants at high allele frequencies, which is consistent with the disruption of RBP target sites having a major impact on fitness, leading to negative selection.
They next investigated the contribution of variants involved in RBP dysregulation to psychiatric disorder heritability. They focused on GWAS from five well-established polygenic psychiatric disorders: attention deficit hyperactivity disorder, autism spectrum disorder, bipolar disorder, major depression, and schizophrenia. The researchers observed significantly elevated levels of RBP dysregulation effect size estimates across all five psychiatric disorders, with 304 cases where target site dysregulation for specific RBPs had a significant effect on psychiatric disorder risk.
"These results indicate that risk variants for psychiatric disorders are extensively spread across RBP target regulatory networks and biochemically underlie the polygenic architecture of mental disorders," the authors wrote. "In particular, we observed significantly larger per-SNP heritability effect sizes for RBPs with dysregulated target sites that are also differentially expressed in the developing human brain."
Further, they found new associations between disrupted RBP target sites and RBPs that had previously been associated with disease. Overall, the data suggested that psychiatric disease risk is significantly linked with perturbations not only of RBPs but also the dysregulation of their targets. Additionally, the researchers found significant dysregulation effects across diverse biochemical regulatory categories of RBPs.
"Methodologically, we demonstrated that deep learning inference of genome-wide molecular effects allows us to estimate major modes of biochemical perturbation and their contribution to disease. We found that splicing disruption is the tip of the iceberg since widespread psychiatric disease risk is associated with RBPs that regulate processes across the life of the RNA," the authors concluded. "This resource, capturing the entire spectrum of common to ultrarare variants, should provide the means to interrogate RBP-derived human diseases at an unprecedented scale."