NEW YORK (GenomeWeb) – Members of the PsychENCODE Consortium have explored the role of non-coding risk variants in regulatory processes that influence brain function and the development of neuropsychiatric conditions such schizophrenia, bipolar disorder, or autism spectrum disorder (ASD) — results they reported in a collection of papers published in Science, Science Translational Medicine, and Science Advances.
Along with psychiatric genomics studies looking at brain development, gene expression, and regulation, for example, teams within the PsychENCODE consortium explored cellular mechanisms behind specific conditions and did disease-specific analyses that incorporated risk variants identified in the past.
In one of the reports in Science, for example, researchers from Yale University, the Allen Institute for Brain Science, and elsewhere brought together messenger RNA sequence, small RNA sequence, single-cell or single-nuclei RNA-seq, chromatin immunoprecipitation sequence, array-based DNA methylation profile, and/or SNP data in hundreds of samples for a functional analysis focused on brain development and neuropsychiatric disease risk.
These included brain tissues or cell types representing as many as 16 brain regions in samples from seven prenatal brains spanning five to 27 weeks after conception, and post-natal brain samples from four individuals ranging in age from newborn up to 64 years old.
From these data, for example, the team was able to start tracking the expression and regulation of brain development- and neuropsychiatric disease-related genes within specific brain regions or cell types across human development and adulthood.
"The presence of these multiple data modalities in a unified resource, and largely from the same tissue samples, allows the integration of information spanning prenatal and post-natal human brain development," authors of that analysis wrote, concluding that their results so far suggest that the resource has potential for doing "integrated analysis for the understanding of brain development and function, and for the rapid interpretation of findings from neuropsychiatric genomics."
For another of the studies published in Science, a team of researchers from Yale University, State University of New York, the University of California at Los Angeles, and elsewhere generated brain enhancer, chromatin interaction, topologically associating domain, and other data for brain samples from 1,866 adult individuals. Together with a deep learning model called the "Deep Structured Phenotype Network" (DSPN), these data made it possible to identify expression quantitative trait loci and map out gene regulatory networks.
Bringing in variants implicated in schizophrenia and other conditions for prior genome-wide association studies, the researchers identified new and known risk gene candidates. They also explored cell types contributing to such conditions based on single-cell RNA sequence data for excitatory neurons, inhibitory neurons and other brain cell types.
"[I]n the future, we can envision how our DSPN approach can be extended to modeling additional intermediate phenotypes," the authors wrote, noting that DSPN is expected to "improve accuracy mainly for complex traits with a highly polygenic architecture, but not necessarily for traits that are strongly determined by only a few variants, such as Mendelian disorders, or are closely correlated with population structure, such as ethnicity."
Using genotype and RNA-seq data for 1,695 individuals, including individuals with ASD, schizophrenia, or bipolar disorders, researchers at UCLA's David Geffen School of Medicine, the Icahn School of Medicine at Mount Sinai, and elsewhere identified transcript splicing and brain expression shifts that appeared to mark each of the conditions — work presented in another Science paper.
"This transcriptome-wide characterization of the molecular pathology across three major psychiatric disorders provides a comprehensive resource for mechanistic insight and therapeutic development," authors of that study wrote.
For a study in Science Translational Medicine, meanwhile, a Central South University-led team used RNA-seq data for 394 post-mortem brain samples from healthy individuals, individuals with schizophrenia, or individuals with bipolar disorder to put together psychiatric disease-related co-expression networks that made up a larger module with a bipolar disease-associated transcription factor gene called POU3F2 in a central position.
"[W]e identified a brain gene expression module that was enriched for rare coding variants in genes associated with schizophrenia and that contained the putative bipolar disorder gene POU3F2," the authors wrote, explaining that the latter transcription factor gene "may be a key regulator of gene expression in this disease-associated gene co-expression module."
For another paper in Science Translational Medicine, a Central South University-led team described a long non-coding RNA suspected of regulating genes containing rare schizophrenia-associated copy number variants, while researchers from the University of Southern California and elsewhere reported in Science Advances on mapped three-dimensional epigenomic features in olfactory neuroepithelium cells from 63 living individuals.
In the remaining studies, PsychENCODE consortium researchers compared transcriptome patterns over time and across brain regions in developing human and macaque brains; explored expression modules in organoid models of cerebral cortical development produced with induced pluripotent stem cells; searched for cell type-specific chromosomal conformation and interaction signatures that occur in the presence of schizophrenia-associated variants; and described non-coding de novo promoter alterations involved in ASD.