reimagined-octo-funicular

Requirements -Python -Biopython -HMMER3 -MEME-Suite -Pfam-A database

Instructions for running lanthipeptide genome mining pipeline:

Note: replace text between < > with what is appropriate for you

Preparing genbank files

1. Download genbank and nucleotide fasta files from NCBI
2. Move file prepare_genbank/prepare-genbank-files.py to directory with the genbank and nucleotide fasta files
3. Edit the range in prepare-genbank-files.py to reflect the number of files you are preparing
4. Execute prepare-genbank-files.py
5. Move file prepare_genbank/index_gb.py to the directory with the genbank files
6. Execute index_gb.py

Extracting potential lanthipeptide biosynthetic gene clusters

1. Edit get_lanC.py so fh = open(<list-of-lanC-like-protein-accession-numbers.txt>) and fh2 = open(<directory_with_genbank_files>/ref_seq-acc.csv')
2. Execute get_lanC.py
3. Edit lanthipeptide_gb_mining.py so variables at the beginning of file match your computer
4. Execute lanthipeptide_gb_mining.py
5. Execute split_classes.py (may need to alter Pfam family IDs if versions are different in your Pfam database)
6. Edit pull-classes.py so fh1 = open() and fh2 = open('lanC_like-acc.csv')
7. Execute pull-classes.py, note results are put to stdout
8. Edit pull-classes.py so fh2 = open('lanC-like-peps.csv')
9. Execute pull-classes.py, note results are put to stdout

Identifying potential precursor peptides in class I clusters

1. Edit make-fasta.py so fh = open()
2. Execute make-fasta.py, note results are put to stdout
3. Execute fimo classI_leader-meme.txt
4. Copy fimo.tsv from created directory to current directory
5. Execute get_params_classI.py <classI-peps-params.csv>
6. Edit select-features.py so fh2 = open(<classI-peps-params.csv>)
7. Execute select-features.py
8. Execute classI_svm_classify.py
9. Execute fimo-compile.py
10. Open classI-peps-params.csv with excel
11. Delete the columns with the amino acid pairs feature columns
12. Open classI-peps-fimo-scores.csv with excel
13. Copy fimo scores and paste into the params file
14. Open classI-peps-classification.csv
15. Copy classifications and paste into params file
16. Calculate score for peptides (+5 for svm classification, +5 for match with leader meme, +2 for core pI less than 9, +2 for 2 or more Cys in core, +1 if leader ends in GG)
17. If peptide score is greater than 10, it is predicted to be a precursor peptide

Identifying potential precursor peptides in class II clusters

1. Edit make-fasta.py so fh = open()
2. Execute make-fasta.py, note results are put to stdout
3. Execute fimo classII_leader-meme_20190503.txt
4. Copy fimo.tsv from created directory to current directory
5. Execute hmmscan -o leader-hmm.out --tblout leader-hmm.tbl classII-precursors.hmm classII-peps.fa
6. Execute get_params_classII.py <classII-peps-params.csv>
7. Edit select-features.py so fh2 = open(<classII-peps-params.csv>)
8. Execute select-features.py
9. Execute classII_svm_classify.py
10. Execute fimo-compile.py
11. Execute hmmer-compile.py
12. Open classI-peps-params.csv with excel
13. Delete the columns with the amino acid pairs feature columns
14. Open classI-peps-fimo-scores.csv with excel
15. Copy fimo scores and paste into the params file
16. Open classI-peps-classification.csv
17. Copy classifications and paste into params file
18. Open classII-peps-HMM.csv in excel
19. Copy HMM scores into params file
20. Calculate score for peptides (+5 for svm classification, +5 for match with leader meme, +5 for match with precursor HMM, +2 for 2 or more Cys in core, +4 if core ends in KRC)
21. If peptide score is greater than 10, it is predicted to be a precursor peptide