Definition of the organization of the ABCA4 transcriptional unit by meta-analysis of transcriptome data

Abstract

Purpose : To gain insight into the genomic organization and transcript composition of ABCA4, the gene responsible for Stargardt disease type 1, using publicly available human retina RNA-Seq datasets.

Methods : A total of 177 bulk RNA-Seq human retina data from non-visually impaired post-mortem donors were retrieved from publicly available expression databases (Pinelli et al., PMID:27235414 and Ratnapriya et al., PMID:30742112). We re-analysed the whole dataset using an ad-hoc designed pipeline. We removed samples with reported RNA integrity number (RIN) values lower than 5.0. Reads were then aligned and mapped to the GRCh38 release of the human genome. Samples with less than 10 million reads mapped and/or less than 70% of reads aligned to the reference genome were discarded. RNA-Seq alignments were assembled to generate an Observed Transcriptome allowing us to create a single set of assembled transcripts and to identify putative novel transcripts. To identify more abundant transcripts, we quantified transcript expression levels by scaling TPM abundance estimates per sample (scaled TPM). We selected ABCA4 transcripts with a median value higher than 50 scaled TPM counts.

Results : After quality control evaluation, we analysed a total of 161 bulk RNA-Seq samples. We focused our analysis of the ABCA4 genomic region, and identified 26 different ABCA4 transcripts, 14 of which are novel. The latter ones are the result of several partial intron retentions, exon skipping and extension along with a few putative novel exon additions.

Conclusions : This is, to the best of our knowledge, the most comprehensive and extended meta-analysis of the ABCA4 locus carried out relying on RNA-Seq data. Our work yielded a reliable expression quantification of the ABCA4 transcripts in the human mature retina, including 16 putatively novel ones, and paves the way towards a better understanding on the organization of this transcriptional unit and on the molecular mechanisms underlying ABCA4-related diseases.

This is a 2021 ARVO Annual Meeting abstract.