Mussismilia hispida enriched metatranscriptomic gene models

The coral-metaorganism transcriptomes response from T3 was explored in order to understand the mechanisms associated with the different phenotypes observed at the end of the experiment. RNA samples were extracted from both the BMC and placebo treatment from the high-temperature (30 °C) and ambient (26 °C) experiment for sequence analysis. Transcriptomic sequencing was performed using Illumina HiSeq 6000 platform (Illumina Inc., San Diego, CA, USA). Transcriptomic reads were quality-controlled using KneadData v0.7.4 with the GRCh38.p13 human genome as a reference for potential decontamination yielding between 14,372,271 and 74,483,759 paired-end (PE) reads at 2×150 bp per sample after quality trimming and filtering. De-novo transcriptomes were co-assembled using rnaSPAdes v3.13.1 (71), producing 520,555 representative transcripts from 496,603 putative genes; genes were estimated by rnaSPAdes. We used TransDecoder v5.5.0 (72) for gene modeling in a multistep process to minimize false positives. In particular, we used the following procedure: i) TransDecoder.LongOrfs, with transcript-to-gene mappings assigned by rnaSPAdes, to generate putative open reading frames (ORFs); ii) hmmscan (hmmer v3.3.1 suite) (73) to identify protein domains using the PFAM v33.1 and TIGRFAM v15.0 databases; iii) Diamond v0.9.30.131 (74) blastp against all Scleractinia (stony corals) proteomes available in NCBI (GCA_002571385.1, GCF_002042975.1, GCA_003704095.1, GCF_004143615.1, GCF_002571385.1, GCF_003704095.1, and GCF_000222465.1); and iv) TransDecoder.Predict with the putative ORFs from (i), the protein domains from (ii), and the alignments from (iii) using the --single_best_only argument. This procedure generated a single ORF per transcript to yield 130,183 ORFs from 114,118 genes. High-quality genes were annotated by using Diamond’s blastp against NCBI’s nr database (v2020.04.01), and taxonomic lineages were extrapolated from NCBItaxid using the get_taxonomy_lineage_from_identifier function from soothsayer v2020.08.24 (https://github.com/jolespin/soothsayer) with ete3 backend (75). PhyloDB v1.076 was used for additional annotations such as KEGG ortholog assignments. <br>