Reference Transcriptome Landscape of Adult Human Retina

Abstract

Purpose : Given the magnitude of retinal degenerative diseases, much effort is invested to find genetic disease variants; yet, a major challenge remains to better understand the biological mechanisms involved. In this study, we performed RNA-seq on 94 human adult retina to characterize in detail the known and novel transcripts of the retinal transcriptome and to understand the gene regulatory network in healthy retina.

Methods : Retina from 45 females and 49 males without retinal disease were procured. RNA was isolated using TRIzol^® and libraries were made with the TruSeq^® Stranded mRNA Library Preparation Kit. The Illumina Hiseq 2500 platform was used for paired-end RNA-seq. Quality control was evaluated using FastQC and RSeQC. Reads were trimmed for adapters and low quality using Trimmomatic, and then aligned to the Ensembl 38.5 genome with STAR 2-pass. RSEM was used to estimate gene- and transcript expression levels such that low read counts (0 CPM across all samples) were removed and stricter filtering (at least 1 CPM in at least 25% of samples) was applied for further comparative analyses. Putative novel transcripts were identified using Cufflinks RABT assembly, TransDecoder, HMMER, and CPAT.

Results : RNA-seq generated 2.2 billion reads (mean 23.5 million reads per sample), of which 94% mapped to the reference genome. 29,422 (64%) of all annotated genes observed were expressed in the retina while more stringent analysis identified 14,025 protein-coding and 1,675 noncoding genes (34% of all annotated genes observed). DAVID analysis of the top 1,000 most highly expressed genes yielded 9 vision-related clusters (total 171); the second cluster related to visual perception while the fourth cluster related to phototransduction, with enrichment scores of 16.12 and 9.76, respectively. 4,015 putative novel isoforms and unknown transcripts were expressed in the retina, including 392 novel lnc RNA. Tissue-specific expression and alternative splicing analyses are underway.

Conclusions : Our study expands on past studies and demonstrates unique aspects of gene expression that can serve as a valuable resource to further understand the transcriptome changes and global molecular processes that define the retina in health and disease.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.