Fig. 1

Data Processing Pipeline. On the far left are the steps taken to assemble the datasets in this manuscript. Pre and Post refer to two different custom software that manage the data. Explanations of each step are written in the figure. The diagram on the right shows how different sequence data are processed, and how protein domain metadata is extracted and processed. GBK files are GenBank format, FNA files are nucleotide FastA files, FAA files are amino acid FastA files. Gene metadata includes the name, accession, and genomic coordinates of a gene or open reading frame. Domain metadata includes name, clan, E-value, and genomic coordinates of a protein domain. The de-overlap process (dagger) is shown in the lower panel. This illustrates how the HMMER3 identified domains are curated to filter out duplicate domains that have been over-identified due to the windowing approach. The E-value is listed after an example domain (showing an example clan). The highlighted domain is compared to each overlapping domain to decide on removal of the overlapping domain based on percentage overlap, E-value, and clan. The domains with green checks would be retained and the others would be removed. 45% and 33% overlapping thresholds are displayed.