Orbicella faveolata and O. franksi coral metagenome assemblies from the Lower Florida Keys region of Florida, USA

The enclosed files include mostly Orbicella faveolata and three Orbicella franksi coral metagenome assemblies collected from the Lower Keys in Florida’s Coral Reef, USA. Metadata for the files is included in this repository. Apparently healthy coral tissue cores were collected between May 28 and June 21, 2021. The DNA was extracted from the host and associated microorganisms and sequenced in a paired-end 150 bp format on an Illumina NovaSeq. Trimming and quality filtering of DNA sequence reads proceeded, followed by host and photoendosymbiotic dinoflagellate DNA removal. The host-cleaned reads were assembled individually by coral sample into longer contigs using MegaHit v1.1.4. The “Assembly_Fastas” zipped file contains 41 metagenome assemblies from the individual Orbicella faveolata corals and 3 assemblies from the individual Orbicella franksi colonies for a total of 44 assemblies. In addition, these assemblies were annotated with eggnog-mapper v2.1.6 to generate both predicted gene regions and annotation output files. The “Predicted_Gene_Fastas” zipped file contains nucleotide fasta files of the predicted gene regions for all 44 coral metagenome assemblies. The fasta header of each gene includes the contig ID it originated from in the associated “Assembly_Fasta”. The “Predicted_Gene_Annotations” zipped file contains either .csv or .xlsx files with the eggnog-mapper-based annotations. These files contain a “query contig” that corresponds to the contig ID in the fasta header of the “Predicted_Gene_Fasta”. 

In addition to individual assemblies, a co-assembly was generated that included all 41 Orbicella faveolata coral samples. Prior to co-assembly, further removal of eukaryotic DNA proceeded by splitting the indiviudual assemblies into eukaryotic and prokaryotic content with the program EukRep v0.6.7, followed by mapping of the host-clean reads to the eukaryotic DNA to remove them. The eukaryote-clean reads from all 41 corals were input into MegaHit to generate a co-assembly. The co-assembly is included (FLK_OFAV_MG_coassembly_final.contigs.fa). Predicted genes from the co-assembly were generated with Prodigal v2.6.3 and the nucleotide fasta of the output is included in this repository (FLK_OFAV_MG_pred.fna). Like with the indiviudal assemblies, eggnog-mapper was used to generate annotations of the predicted genes from Prodigal (FLK_OFAV_MG.emapper.annotations.xlsx).  Additionally, the abundance of each predicted gene was generated using Salmon to map the eukaryote-clean reads to the predicted genes. The number of reads (counts) for each gene across each coral sample were aggregated as integers into one table and included in this repository (FLK_OFAV_MG_pred_NumReads.tsv).  

These data were processed and generated by Julie Meyer’s Lab at the University of Florida, using funding from the Florida Department of Environmental Protection.