Skip to content

v.0.2.0
Compare
Choose a tag to compare
@Darcy220606 Darcy220606 released this 20 Feb 16:07
· 5 commits to main since this release
v.0.2.0
ce68c7b

AMPcombi v.0.2.0:
This update was done with extensive scientific advice from Dr. Rosa Herbst.

Changes made include:

  • Change the output extension from .csv to .tsv
  • Added a new feature to estimate the isoelectric point, mwt, structure fraction and hydrophobicity
  • The DRAMP db .tsv was filtered to retain only necessary columns
  • Added the tool AMPtransformer
  • Added the tool AMPgram
  • Fixed hmmsearch to now parse both single and multi models
  • Fixed the dependencies in the environment.yml file
  • Added a log file for each sample, besides the main log for the complete summary. Beneficial when used in pipelines like nf-core/funcscan
  • Added the parameter '--sample_metadata' an optional flag to add a file with sample metadata
  • Added the parameter '--contig_metadata' an optional flag to add a file with contig metadata
  • Removed tooldictionary parameter '--tooldict' and added a parameter for each tool instead '--_file' to assign the extension of the files per tool
  • Added the parameter --hmm_evalue to filter the hmm results in the table based on E-value
  • Added the parameter --aminoacid_length to retain hits with a specific length
  • Added the parameter --db_evalue to filter out classification hits below specific diamond E-values
  • Renamed the parameter --cutoff to --amp_cutoff
  • Added the parameter '--gbk' to add gbk files as input for parsing to include contig ids and other important values
  • Added the parameter '--window_size_stop_codon' that adjusts the window size to look for stop codons
  • Added the parameter '--remove_stop_codons' that removes hits with no stop codons found in the vicinity
  • Added the parameter '--window_size_transporter' that adjusts the window size to look for 'transporter' proteins
  • Added a new output directory per sample that holds the gbk files for the contigs with hits
  • Added the submodule to cluster the AMP hits with the parameters to modify the clustering using MMSeqs2 for e.g., '--cluster_cov_mode' '--cluster_mode' '--cluster_coverage' '--cluster_seq_id' '--cluster_sensitivity' '--cluster_remove_singletons' '--cluster_retain_label' '--cluster_min_member'
  • Added the submodule to predict signaling peptide using SignalP-6.0h
  • Removed the HTML output for the complete summary and replaced it with shiny for python app that can be viewed using the commandline
  • Updated AMPcombi to use subcommands for ease of use with standardized and portable piplines
Assets 2
Loading