Surveying alternative transcripts of disease-associated genes and their cell type specificity - a case study with ANK3.

The ANK3 gene has been conclusively linked to bipolar disorder, but its role in disease etiology remains unclear: both the specific transcript and its expressing cell type remain unclear. This is a near-universal problem for genes associated with polygenic neuropsychiatric disorders. One of the causes of this unfortunate stasis is that human transcript databases have an incomplete list of alternative isoforms of genes and that it is technically challenging to identify the cell-types in which they are expressed. Here, we address this knowledge gap for ANK3. ANK3 is a useful case study because of its rich alternative splicing and important neurobiological role in neurons and oligodendrocytes. We used Oxford nanopore sequencing of full-length transcripts to identify all human and mouse ANK3 exons. Then, we isolate and sort brain nuclei and use droplet-digital PCR to measure cell-type specific expression of alternatively spliced exons. We confirm that RefSeq transcript annotations of ANK3 lack many of the alternatively spliced exons and we show that, despite being lowly expressed, these exons are highly evolutionarily conserved and have strong cell type specificity. Elucidating psychiatric disorder etiology will require an overview of the transcripts of all the disorder-associated genes and the cell types in which these transcripts are expressed. This work on ANK3 illustrates the extent to which such information is currently lacking, ways in which the information can be acquired, and the technical challenges that are encountered in the process.