search

griffithlab / pVACtools

http://www.pvactools.org
BSD 3-Clause Clear License
137 stars 59 forks source link

IEDB_MHC_II-3.1.9 installation error #1019

Closed serge2016 closed 10 months ago

serge2016 commented 1 year ago

Installation Type

Standalone

pVACtools Version / Docker Image

-

Python Version

3.7.5

Operating System

ubuntu 18.04

Describe the bug

Hello! I know, that you don't support IEDB, but maybe you know anything about this... Here (https://github.com/griffithlab/docker-pvactools/blob/master/Dockerfile#L41) you use version 3.1.6 and not 3.1.9. Why? Is this due to broken self test in the last version?

MHCII_VERSION="3.1.9"
wget -q "https://downloads.iedb.org/tools/mhcii/${MHCII_VERSION}/IEDB_MHC_II-${MHCII_VERSION}.tar.gz" -O "$IEDB_PATH/IEDB_MHC_II-${MHCII_VERSION}.tar.gz"
tar -C "$IEDB_PATH" -xzf "$IEDB_PATH/IEDB_MHC_II-${MHCII_VERSION}.tar.gz"
cd "$IEDB_PATH/mhc_ii"
python3 ./configure.py

All prerequisites found!
run MHCII tests...
...
FAIL: test_recommended (__main__.Test_test)
...
#3.1.9
root@e90a37c0742c:/soft/IEDB/mhc_ii# python mhc_II_binding.py IEDB_recommended HLA-DRB1*03:01 test.fasta
allele  seq_num start   end     length  core_peptide    peptide score   percentile_rank                 
HLA-DRB1*03:01  1       3       17      15      LEGVSGATW       NRDFLEGVSGATWVD 0.001547        79.0    
HLA-DRB1*03:01  1       8       22      15      TWVDLVLEG       EGVSGATWVDLVLEG 0.001172        83.0    
HLA-DRB1*03:01  1       4       18      15      LEGVSGATW       RDFLEGVSGATWVDL 0.001156        84.0    
HLA-DRB1*03:01  1       2       16      15      LEGVSGATW       SNRDFLEGVSGATWV 0.000854        87.0    
HLA-DRB1*03:01  1       1       15      15      LEGVSGATW       MSNRDFLEGVSGATW 0.000664        90.0    
HLA-DRB1*03:01  1       5       19      15      LEGVSGATW       DFLEGVSGATWVDLV 0.000591        91.0    
HLA-DRB1*03:01  1       7       21      15      VSGATWVDL       LEGVSGATWVDLVLE 0.000448        93.0    
HLA-DRB1*03:01  1       6       20      15      VSGATWVDL       FLEGVSGATWVDLVL 0.000229        97.0

Expected (according to the test):

allele  seq_num start   end     length  method  peptide consensus_percentile_rank       adjusted_consensus_percentile_rank      comblib_core    comblib_score   comblib_rank    adjusted_comblib_rank   smm_align_core  smm_align_ic50  smm_align_rank  adjusted_smm_align_rank nn_align_core   nn_align_ic50   nn_align_rank
adjusted_nn_align_rank  netmhciipan_core        netmhciipan_ic50        netmhciipan_rank        adjusted_netmhciipan_rank       sturniolo_core  sturniolo_score sturniolo_rank  adjusted_sturniolo_rank
HLA-DRB1*03:01  1       8       22      15      SMM-NN-Sturniolo        EGVSGATWVDLVLEG 59.0    59.00   -       -       -       -       VSGATWVDL       5932.0  59.0    59.00   TWVDLVLEG       3356.9  50.0    50.00   -       -       -       -       VSGATWVDL       -1.34   88.0    88.00
HLA-DRB1*03:01  1       4       18      15      SMM-NN-Sturniolo        RDFLEGVSGATWVDL 85.0    85.00   -       -       -       -       FLEGVSGAT       8523.0  69.0    69.00   VSGATWVDL       12707.7 85.0    85.00   -       -       -       -       FLEGVSGAT       -1.2    87.0    87.00
HLA-DRB1*03:01  1       5       19      15      SMM-NN-Sturniolo        DFLEGVSGATWVDLV 85.0    85.00   -       -       -       -       FLEGVSGAT       18841.0 85.0    85.00   VSGATWVDL       10634.4 81.0    81.00   -       -       -       -       FLEGVSGAT       -1.2    87.0    87.00
HLA-DRB1*03:01  1       6       20      15      SMM-NN-Sturniolo        FLEGVSGATWVDLVL 85.0    85.00   -       -       -       -       FLEGVSGAT       18363.0 85.0    85.00   VSGATWVDL       8154.8  74.0    74.00   -       -       -       -       FLEGVSGAT       -1.2    87.0    87.00
HLA-DRB1*03:01  1       1       15      15      SMM-NN-Sturniolo        MSNRDFLEGVSGATW 87.0    87.00   -       -       -       -       FLEGVSGAT       9343.0  71.0    71.00   DFLEGVSGA       17372.9 92.0    92.00   -       -       -       -       FLEGVSGAT       -1.2    87.0    87.00
HLA-DRB1*03:01  1       2       16      15      SMM-NN-Sturniolo        SNRDFLEGVSGATWV 87.0    87.00   -       -       -       -       FLEGVSGAT       8986.0  70.0    70.00   DFLEGVSGA       18531.9 93.0    93.00   -       -       -       -       FLEGVSGAT       -1.2    87.0    87.00
HLA-DRB1*03:01  1       3       17      15      SMM-NN-Sturniolo        NRDFLEGVSGATWVD 87.0    87.00   -       -       -       -       FLEGVSGAT       9085.0  70.0    70.00   DFLEGVSGA       18458.2 93.0    93.00   -       -       -       -       FLEGVSGAT       -1.2    87.0    87.00
HLA-DRB1*03:01  1       7       21      15      SMM-NN-Sturniolo        LEGVSGATWVDLVLE 88.0    88.00   -       -       -       -       SGATWVDLV       26621.0 91.0    91.00   VSGATWVDL       5473.3  63.0    63.00   -       -       -       -       VSGATWVDL       -1.34   88.0    88.00

The test itself didn't change since 3.1.6 and the output has changed... And I suppose it should! Updated 25-01-21: Default threshold for WB changed to 5%, and SB to 1% [MN]

How to reproduce this bug

#IEDB/mhc_ii
python3 ./configure.py Test_test.test_recommended

Log output

All prerequisites found!
run MHCII tests...
/usr/lib/python3.7/subprocess.py:883: ResourceWarning: subprocess 2377 is still running
  ResourceWarning, source=self)
ResourceWarning: Enable tracemalloc to get the object allocation traceback
./configure.py:23: ResourceWarning: unclosed file <_io.TextIOWrapper name=3 encoding='UTF-8'>
  result=os.popen(cmd).read()
ResourceWarning: Enable tracemalloc to get the object allocation traceback
F
======================================================================
FAIL: test_recommended (__main__.Test_test)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "./configure.py", line 24, in test_recommended
    self.assertEqual(result, 'allele\tseq_num\tstart\tend\tlength\tmethod\tpeptide\tconsensus_percentile_rank\tadjusted_consensus_percentile_rank\tcomblib_core\tcomblib_score\tcomblib_rank\tadjusted_comblib_rank\tsmm_align_core\tsmm_align_ic50\tsmm_align_rank\tadjusted_smm_align_rank\tnn_align_core\tnn_align_ic50\tnn_align_rank\tadjusted_nn_align_rank\tnetmhciipan_core\tnetmhciipan_ic50\tnetmhciipan_rank\tadjusted_netmhciipan_rank\tsturniolo_core\tsturniolo_score\tsturniolo_rank\tadjusted_sturniolo_rank\nHLA-DRB1*03:01\t1\t8\t22\t15\tSMM-NN-Sturniolo\tEGVSGATWVDLVLEG\t59.0\t59.00\t-\t-\t-\t-\tVSGATWVDL\t5932.0\t59.0\t59.00\tTWVDLVLEG\t3356.9\t50.0\t50.00\t-\t-\t-\t-\tVSGATWVDL\t-1.34\t88.0\t88.00\nHLA-DRB1*03:01\t1\t4\t18\t15\tSMM-NN-Sturniolo\tRDFLEGVSGATWVDL\t85.0\t85.00\t-\t-\t-\t-\tFLEGVSGAT\t8523.0\t69.0\t69.00\tVSGATWVDL\t12707.7\t85.0\t85.00\t-\t-\t-\t-\tFLEGVSGAT\t-1.2\t87.0\t87.00\nHLA-DRB1*03:01\t1\t5\t19\t15\tSMM-NN-Sturniolo\tDFLEGVSGATWVDLV\t85.0\t85.00\t-\t-\t-\t-\tFLEGVSGAT\t18841.0\t85.0\t85.00\tVSGATWVDL\t10634.4\t81.0\t81.00\t-\t-\t-\t-\tFLEGVSGAT\t-1.2\t87.0\t87.00\nHLA-DRB1*03:01\t1\t6\t20\t15\tSMM-NN-Sturniolo\tFLEGVSGATWVDLVL\t85.0\t85.00\t-\t-\t-\t-\tFLEGVSGAT\t18363.0\t85.0\t85.00\tVSGATWVDL\t8154.8\t74.0\t74.00\t-\t-\t-\t-\tFLEGVSGAT\t-1.2\t87.0\t87.00\nHLA-DRB1*03:01\t1\t1\t15\t15\tSMM-NN-Sturniolo\tMSNRDFLEGVSGATW\t87.0\t87.00\t-\t-\t-\t-\tFLEGVSGAT\t9343.0\t71.0\t71.00\tDFLEGVSGA\t17372.9\t92.0\t92.00\t-\t-\t-\t-\tFLEGVSGAT\t-1.2\t87.0\t87.00\nHLA-DRB1*03:01\t1\t2\t16\t15\tSMM-NN-Sturniolo\tSNRDFLEGVSGATWV\t87.0\t87.00\t-\t-\t-\t-\tFLEGVSGAT\t8986.0\t70.0\t70.00\tDFLEGVSGA\t18531.9\t93.0\t93.00\t-\t-\t-\t-\tFLEGVSGAT\t-1.2\t87.0\t87.00\nHLA-DRB1*03:01\t1\t3\t17\t15\tSMM-NN-Sturniolo\tNRDFLEGVSGATWVD\t87.0\t87.00\t-\t-\t-\t-\tFLEGVSGAT\t9085.0\t70.0\t70.00\tDFLEGVSGA\t18458.2\t93.0\t93.00\t-\t-\t-\t-\tFLEGVSGAT\t-1.2\t87.0\t87.00\nHLA-DRB1*03:01\t1\t7\t21\t15\tSMM-NN-Sturniolo\tLEGVSGATWVDLVLE\t88.0\t88.00\t-\t-\t-\t-\tSGATWVDLV\t26621.0\t91.0\t91.00\tVSGATWVDL\t5473.3\t63.0\t63.00\t-\t-\t-\t-\tVSGATWVDL\t-1.34\t88.0\t88.00\n')
AssertionError: 'alle[28 chars]gth\tcore_peptide\tpeptide\tscore\tpercentile_[594 chars].0\n' != 'alle[28 chars]gth\tmethod\tpeptide\tconsensus_percentile_ran[1976 chars]00\n'
Diff is 2429 characters long. Set self.maxDiff to None to see it.

----------------------------------------------------------------------
Ran 1 test in 0.616s

FAILED (failures=1)
serge2016 commented 1 year ago

Same problem with 3.1.8. 3.1.7 and 3.1.6 works fine.

susannasiebert commented 1 year ago

I've never had any problems actually running IEDB even if the tests fail so I have generally been ignoring these failures.

In regards to the Docker file, you will want to look at the version-specific branches (e.g. https://github.com/griffithlab/docker-pvactools/blob/4.0.3/Dockerfile). We're using IEDB MHCII 3.1.8 currently but I will make a note to update to the latest version when we release a new pVACtools version.

serge2016 commented 1 year ago

Thank you, Susanna! I didn't observe these branches. I'll try!

susannasiebert commented 10 months ago

It looks like this issue was resolved. Closing.

serge2016 commented 9 months ago

I confirm! Thank you very much!