Abstract:

Here we present a summary of the generation of dog Single-molecule Real-time Isoform Sequencing (Iso-Seq) data, and its application for transcriptome annotation. Iso-Seq libraries from six tissues were sequenced, and read clean-up was implemented with IsoSeq V3 tools. A high retention rate in demultiplexing and refining steps, as well as a high mapping rate (90%) of Full-length non-concatemer (FLNC) reads, guaranteed enough useful information for subsequent analyses. Finally, 17,766-45,170 clustered transcripts from the six libraries were aligned to the ROS_Cfam_1.0, a male Labrador retriever dog reference genome assembly. In terms of annotation, we identified 14,059 to 26,263 transcripts and 4,985 to 10,457 genes with the sequence data in each of the libraries, with 32 to 68% unique transcripts being expressed. Coding transcripts in each tissue are expressed in a relatively stable manner, from 60-70%. This study provides an improved dog genome annotation resource and offers new insights into tissue-specific transcriptome pattern.