search

rajewsky-lab / polyA

0 stars 1 forks source link

New tests #64

Closed mschilli87 closed 8 years ago

mschilli87 commented 8 years ago

Working on issue #60:

Adding unit tests to ensure that genes listed as single UTR no internal poly(A) sites have

edit:

As the single UTR test above failed, I'll add additional tests to figure out what is going on.

These test will ensure that genes listed as single UTR no internal poly(A) sites have

codecov-io commented 8 years ago

Current coverage is 72.64%

Merging #64 into new_tests will increase coverage by 0.56%

@@           new_tests        #64   diff @@
===========================================
  Files              3          3          
  Lines            308        318    +10   
  Methods            0          0          
  Messages           0          0          
  Branches           0          0          
===========================================
+ Hits             222        231     +9   
- Misses            86         87     +1   
  Partials           0          0

Powered by Codecov. Last updated by 139bb24...04a2d5c

mschilli87 commented 8 years ago

There are two issues:

  1. The single UTR test _failed!_
  2. Nevertheless, unit_test.py still returns 0, resulting in a positive build status.

I'll investigate the first in more detail & opened #65 for the latter.

mschilli87 commented 8 years ago

What we know so far:

nukappa commented 8 years ago

thanks mschilli, it could very well be that the problem is in the list itself (don't remember how and when i derived it). i'll write a small function extracting them directly from the annotation file.

nukappa commented 8 years ago

quick and dirty: how about using these one-liners to create new files? when i'll have more time i'll write them in python. -- extract_single_utr_genes.sh zcat test_data/Homo_sapiens.GRCh38.84_chr9.gtf.gz | awk '$3 == "three_prime_utr" {print $18}' | sort | uniq -c | awk '$1 == 1 {print $2}' | cut -c2- | sed 's/..$//' -- extract_genes_with_pAi.sh awk '{print $4}' test_data/output/pAi_gene.bed | uniq | sort | uniq -- extract all single UTR genes WITHOUT pAi comm -23 <(extract_single_utr_genes.sh) <(extract_genes_with_pAi.sh)

mschilli87 commented 8 years ago

I tried to set this up & it might work but this will require building test_data/output/pAi_gene.bed for every unit test which will also need the corresponding FASTA file first. Do we really want to go there or just update the list manually once now & keep it in the repo for the time being?

nukappa commented 8 years ago

no no no! we don't want to go there! my motivation was that you inspect the one-liners and if you find them ok just update the list manually and add it to the repo :)

mschilli87 commented 8 years ago

Once https://github.com/rajewsky-lab/polyA/pull/69/commits/663a9baf81e0f36f86c4f74947ab8b94c99d5b5d is merged to master, I'll merge it into here & proceed with updating test_data/single_utr_no_pAi_genes.txt as suggested.

nukappa commented 8 years ago

cool. i merged the bugfix into the master, everything should be fine now.

mschilli87 commented 8 years ago

I merged in https://github.com/rajewsky-lab/polyA/pull/72/commits/021f7ec694bde85f079d90796888206348d6b8b9 and all tests passed now! Thus, pulling this into master will finally fix #60.

nukappa commented 8 years ago

perfect :) should i test and merge or yo uhave more to commit?

mschilli87 commented 8 years ago

I labeled 'test & merge to master' & assigned you. ;)

nukappa commented 8 years ago

haha ok i missed the "being assigned" part. will do asap