Neuropsychiatric disorders are debilitating conditions with a strong genetic component. Genome-wide association studies (GWAS) have identified hundreds of genomic risk loci for these neuropsychiatric disorders [1, 2]. How these risk loci and their associated risk genes contribute to disease risk through altered gene expression and RNA splicing is not well understood [3]. We used long-read sequencing combined with long-range PCR to generate full-length sequence data for risk genes to better understand their expression, splicing and contribution to neuropsychiatric disease risk.
To identify likely causal risk genes from GWAS loci, we collated expression, association and pathway information and selected genes with multiple lines of evidence for schizophrenia, bipolar disorder, depression and autism spectrum disorder. We amplified the entire coding regions of 18 high-confidence risk genes from seven regions of post-mortem human brain from six individuals. Minimap2, Salmon and TAQLoRe were used to align sequences to the human genome, quantify transcripts and identify novel transcripts respectively.
Lengths of amplified coding regions for 18 risk genes ranged from 800 nt to 10.8 kb. Pilot sequencing of 13 risk genes identified 93 transcripts, including 60 annotated protein coding isoforms. TAQLoRe analysis identified 78 novel exons (range: 1 – 18 exons) in 10 of 13 risk genes. Novel isoforms for some genes were abundantly expressed with three novel isoforms contributing to over 25% of MAPT expression.
The outcomes of this study support the use of long-read Nanopore sequencing to identify full-length novel isoforms in different human brain regions. These novel exons and isoforms may be linked with disease and will improve our understanding of neuropsychiatric disorder causation and RNA splicing profiles in the brain.