tleonardi/bedparse

Python module and CLI tool to perform operations on BED files

bedparse's Issues

Add support for BED files containing extra fields

convertChr should support BED files with extra fields (or produce a meaningful error message).

bedparse convert fails on nanocompore bed file

Hi,
I am having issues with using bedparse to convert an ensembl bed file into UCSC. The bed file was made with nanocompore and the alignments were made to gencode.v33_transcripts which contain the transcript version numbers.
When I run the following code on a bed file with the header removed
bedparse convertChr --assembly hg38 --target ucsc WT_v_KO_DRS.2_sig_sites_GMM_logit_pvalue_context_2_thr_0.01.bed
The script fails on the first line of the bed file and I get the following error;

Traceback (most recent call last):
File "/home/samirwatson/miniconda3/envs/guitar/bin/bedparse", line 10, in
sys.exit(main())
File "/home/samirwatson/miniconda3/envs/guitar/lib/python3.9/site-packages/bedparse/bedparse.py", line 250, in main
args.func(args)
File "/home/samirwatson/miniconda3/envs/guitar/lib/python3.9/site-packages/bedparse/bedparse.py", line 121, in convertChr
translatedLine=bedline(line.split('\t')).translateChr(assembly=args.assembly, target=args.target, suppress=args.suppressMissing, ignore=args.allowMissing, patches=args.patches)
File "/home/samirwatson/miniconda3/envs/guitar/lib/python3.9/site-packages/bedparse/bedline.py", line 524, in translateChr
raise BEDexception("The chromosome of transcript "+self.name+" ("+self.chr+") can't be found in the DB.")
bedparse.BEDexception: The chromosome of transcript ENST00000368723.4_ACCTC (chr1) can't be found in the DB.

I have checked the bed file using the validateFormat function and it passed.

Improve support for GTF files other than Ensembl/Gencode

For example, gtf2bed fails with the gtf file from XenBase:
ftp://ftp.xenbase.org/pub/Genomics/JGI/Xenla9.2/XL9_2_GCA.gff3

Add tests for tx2genome

Use dynamically generated CLI docs

While not an issue, you may be interested in using something like sphinxcontrib-programoutput to have your CLI help command generated upon each commit so that it always stays up to date.

The one rub would be that you'd need to switch from markdown to rst, but that should be easy using panda. If you're interested and run into any trouble, I'm happy to help out.

Add community guidelines

As per the JOSS review requirements.

Filtering not working in gtf2bed

Filtering is based on the 'transcript_type' field, but Ensembl GTF has a 'transcript_biotype' field.

Implement gtf2bed filters

When converting a GTF file to BED, allow to filter by gene type.

No stripping of trailing new lined for BED lines of less than 12 fields

This causes beparse join to fail for BED4 format, because the txName contains the newline and can't be matched with the annotation.

tx2genome error when run on BED6

The function throws the following error when used on a BED6 record:

AttributeError: ("'bedline' object has no attribute 'exStarts'", 'occurred at index 0')

Add description in pypi

Add long_description in setup.py as in https://packaging.python.org/tutorials/packaging-projects/#setup-args

Unable to parse BED3 file from Ensembl example

I tried using the BED3 file from the Ensembl example and am currently getting:

(tempenv-19e22026634fb) ~ ❯❯❯ cat test.bed
chr1  213941196  213942363
chr1  213942363  213943530
chr1  213943530  213944697
chr2  158364697  158365864
chr2  158365864  158367031
chr3  127477031  127478198
chr3  127478198  127479365
chr3  127479365  127480532
chr3  127480532  127481699
(tempenv-19e22026634fb) ~ ❯❯❯ bedparse cds test.bed
Traceback (most recent call last):
  File "/Users/BenjaminLee/.virtualenvs/tempenv-19e22026634fb/bin/bedparse", line 10, in <module>
    sys.exit(main())
  File "/Users/BenjaminLee/.virtualenvs/tempenv-19e22026634fb/lib/python3.6/site-packages/bedparse/bedparse.py", line 222, in main
    args.func(args)
  File "/Users/BenjaminLee/.virtualenvs/tempenv-19e22026634fb/lib/python3.6/site-packages/bedparse/bedparse.py", line 40, in cds
    utr=bedline(line.split('\t')).cds(ignoreCDSonly=args.ignoreCDSonly)
  File "/Users/BenjaminLee/.virtualenvs/tempenv-19e22026634fb/lib/python3.6/site-packages/bedparse/bedline.py", line 29, in __init__
    raise BEDexception("Only BED3,4,6,12 are supported. "+self.name+" is neither.")
bedparse.BEDexception: Only BED3,4,6,12 are supported. NoName is neither.

Is this behavior anticipated? Either way, could you provide some example files in the repository for experimentation?

Mistake in documentation for bedparse promoter

The docs says that the subcommand reports the promoters of coding genes, while it prints promoters of all transcripts in the input BED.

Add link to docs at top of Github repo

Not sure what it's called, but next to the description for the repo at the very top, there's a spot for a link. Would be nice to put the link to your docs up there.

See here for an example.

Get rid of print() and pprint()

The print() and pprint() methods should be removed and their functionality implemented in str and repr.
See thread in JOSS review