DiazoDB

Nitrogenase Database

Automated annotation and curation of nitrogenase genes using profile hidden Markov models and conserved residue matching.

Install

Scripts

• 1_align_mafft.sh - Clusters (mmseqs easy-cluster) HMM seed sequences at 90% AAI. Aligns (MAFFT) clustered seeds sequences. Builds HMM profiles and combines into single file. Stores alignmnets in alignments/ and profiles in HMMs/.

• 2_hmmsearch.sh - Runs pHMM search against GTDB R220 all_rep_proteins_aa database (bacteria and archaea). Stores results as individual files in results/{archaea | bacteria} (per GTDB accession number)

• 3_parse_hmm.sh - Calls (1) Parse_hmm_results.py and (2) Parse_hmm_results.py.

• 4_conserved_res.sh - Calls (1) aln_nif_hits.py, (2) conserved-res.py, (3) final-fasta-export.py, and (4) diazoDB-check.py.

• 5_make_trees.sh -

• 6_operon_org.sh

• 7_SSN.sh

• Parse_hmm_results.py - Parse pHMM search results (see below criteria), combines with GTDB taxonomy data, and stores results in hits.feather

  • positive bit score
  • full sequence evalue must be significant (<0.01)
  • best domain evalue should be significant (<0.01)
    • otherwise flagged for manual review to check if it is distant homolog or just short repeats

• Parse_tophits.py

• aln_nif_hits.py

• cluster_pos.py

• conserved-res.py

• final-fasta-export.py

• diazoDB-check.sh

• get-operon.py

• operon-org-plot.py