Long-Read Transcriptome of Equine Bronchoalveolar Cells

Abstract

We used Pacific Biosciences long-read isoform sequencing to generate full-length transcript sequences in equine bronchoalveolar lavage fluid (BALF) cells. Our dataset consisted of 313,563 HiFi reads comprising 805 Mb of polished sequence information. The resulting equine BALF transcriptome consisted of 14,234 full-length transcript isoforms originating from 7017 unique genes. These genes consisted of 6880 previously annotated genes and 137 novel genes. We identified 3428 novel transcripts in addition to 10,806 previously known transcripts. These included transcripts absent from existing genome annotations, transcripts mapping to putative novel (unannotated) genes and fusion transcripts incorporating exons from multiple genes. We provide transcript-level data for equine BALF cells as a resource to the scientific community.

Keywords: Equus caballus; Iso-Seq; annotation; asthma; bioinformatics; horse; lung.

Figure 1

(A) Schematic overview of the bioinformatics workflow used to generate full-length transcript annotations. (B) Length comparison between reference transcript length and Iso-Seq data.

Figure 2

(A). UCSC genome browser track of CD163 in equine BALF cells. (B). UCSC genome browser track showing the transcripts of the fusion gene MPDU1_CD68 (green track). These transcripts cover both the genes MPDU1 and CD68. Their biological significance is unknown at this time.

Figure 3

(A). Binned distances between the Iso-Seq transcription start sites (TSS) and the corresponding annotated TSS (FSM transcripts only). (B). Binned distances between the Iso-Seq transcript poly(A) site and the corresponding annotated poly(A) site (FSM transcripts only). (C). In contrast to FSM reads, most ISM reads do not extend beyond the 5′-UTRs of known transcripts. (D). FSM and ISM reads extending beyond the 3′-UTRs of known transcripts. TSS: transcriptional start site. TTS: transcriptional termination site.

Figure 4

(A). Top ten splice sites identified in the known (previously annotated) and novel transcript isoforms. (B). Distribution of splice junction types across transcript structural classes.