schultzm / havic Goto Github PK

Add distance method as an option to argparse, to let the user choose which pairwise distance algorithm to use in clusterpicker. Version currently uses 'valid'. Need to test the 'ambiguity' method to understand whether e.g., A will be called a difference to M (either A or C).

get the r dependency check to work properly

Important to get this finished ASAP.

make branch support cutoffs accessible as an option from argparse

I now have havic running on a test dataset for measles, so theoretically should work for any single contig reference genome and target region. Rename havic to something more generic and more accurately descriptive.

Here is an example:

https://www.freecodecamp.org/news/a-beginners-guide-to-git-what-is-a-changelog-and-how-to-generate-it/

The docs say to just rune install.sh ?

It would be good to make that script safer.
Look up definsive bash programming

1. add #!/bin/bash line
2. add set -e -u so it stops if one of the commands fail
   eg. conda or R is not installed

It should probably warn that it is about to make 100s of files?

the ref amplicon is hard-coded. make it an option

Please help me to convert the ruffus pipeline to snakemake. I will make a list of the step inputs and outputs and send to you @andersgs

https://docs.github.com/en/github/building-a-strong-community/setting-guidelines-for-repository-contributors

https://docs.github.com/en/github/building-a-strong-community/adding-a-code-of-conduct-to-your-project

• switch over to python3.8
• Check correctness of minimap2 implementation (is it the best choice here, based on end-soft clips and IUPAC handling or masking)
• add an option for fofn as inputs as opposed to having to list the input files on the command line, or reset the input file to a single file as opposed to allowing a list of input files
• switch trim header list to file of trim headers
• upgrade to iqtree2
• check IQtree settings to ensure appropriateness of parameters given version2 release of IQtree
• check appropriateness of ClusterPicker as the tool for cluster-picking: suggest here use of phydelity instead https://github.com/alvinxhan/Phydelity
• CI using CircleCI
• print to stdout any sequence headers that have been modified by HAVIC (e.g., replacements of non A-z|0-9 with _)
• if ClusterPicker has no .jar extension, drop the java -jar part from the run cmd
  install via conda is new method

When no trim seqs are specified, the list search breaks:

Traceback (most recent call last):
  File "/home/schultzm/.local/bin/havic", line 10, in <module>
    sys.exit(main())
  File "/home/schultzm/.local/lib/python3.7/site-packages/havic/__main__.py", line 197, in main
    args.iqtree_threads)
  File "/home/schultzm/.local/lib/python3.7/site-packages/havic/utils/pipeline_runner.py", line 71, in __init__
    .replace("(", "").replace(")", "").rstrip()) for i in trim_seqs]
TypeError: 'NoneType' object is not iterable

Needed for JOSS submission

locking ncpus to 2 and 4 stops efficient run of IQtree with longer alignments

.sqlite and NFS are incompatible. Move the .sqlite ruffus database to somewhere not on the NFS (/tmp)

Moved demo test suite over to unittest. Added six additional unittests.