Bad id problem from bmtagger while running kneaddata v.0.7.10

Hello,

I’m using kneaddatav0.7.10 (installed via pip) and I’m having issue with bmtagger. The bmtagger utilities were downloaded from here.

Here is my error message:

10/30/2020 01:28:33 PM - kneaddata.utilities - CRITICAL: Error executing: /home/hfan/build/bmtagger/bmtagger.sh -q 1 -1 sample.R1_kneaddata.repeats.removed.1.fastq -T sample.R1_kneaddata_paired_temp_u8312fsj --extract -b Homo_sapiens_BMT
agger_v0.1/Homo_sapiens_assembly19.fasta.bitmask -x Homo_sapiens_BMTagger_v0.1/Homo_sapiens_a
ssembly19.fasta.srprism -o sample.R1_kneaddata_paired_Homo_sapiens_assem
bly19.fasta_bmtagger -2 sample.R1_kneaddata.repeats.removed.2.fastq

Error message returned from bmtagger :
Info: no ./bmtagger.conf found
Using following programs:
/home/hfan/build/bmtagger/bmfilter
/home/hfan/build/bmtagger/srprism
/home/hfan/miniconda3/bin/blastn
/home/hfan/build/bmtagger/extract_fullseq
MAIN SCRIPT IS /home/hfan/build/bmtagger/bmtagger.sh (PID=4390)
RUNNING bmfilter
* Attaching Homo_sapiens_BMTagger_v0.1/Homo_sapiens_assembly19.fasta.bitmask (pattern = 0b111
111111111111111 of len 18 using 36 bits)
* Notice: creating CFastqFileReader( 1, 0, sample.R1_kneaddata.repeats.r
emoved.1.fastq, sample.R1_kneaddata.repeats.removed.2.fastq)
* Info: Created sample.R1_kneaddata_paired_temp_u8312fsj/bmtagger.160403
5684.Cosette.4390.tag
terminate called after throwing an instance of 'std::runtime_error'
  what():  
runtime_error ???:0 [0x45913a] offset=0x0
runtime_error ???:0 [0x4591fe] offset=0x0
runtime_error ???:0 [0x4316d2] offset=0x0
runtime_error ???:0 [0x433a79] offset=0x0
runtime_error ???:0 [0x40368b] offset=0x0
runtime_error ???:0 [0x5a1450] offset=0x0
runtime_error ???:0 [0x4001b9] offset=0x0


In file cshortreader.cpp line 189 runtime_error: Error: bad read id in file 
sample.R1_kneaddata.repeats.removed.2.fastq: expected A00204:533:HC5CLDSXY:1:1101:16043:1000:N:0:CTAGTTCG+GAAGCCTA#0/1, got 
A00204:533:HC5CLDSXY:1:1101:10203:1000:N:0:CTAGTTCG+GAAGCCTA#0/2

/home/hfan/build/bmtagger/bmtagger.sh: line 179:  4398 Aborted                 (core dumped) $BMFILTER $accession $input1 $input
2 -q $quality $bmfiles $spotId_only -o "$TMPDIR"/bmtagger.$tmpstr $bmoptions

real	0m29.054s
user	0m0.013s
sys	0m3.514s
FAILED: bmfilter with rc=134

As you can see it seems like the problem it’s complaining is:

In file cshortreader.cpp line 189 runtime_error: Error: bad read id in file 
sample.R1_kneaddata.repeats.removed.2.fastq: expected A00204:533:HC5CLDSXY:1:1101:16043:1000:N:0:CTAGTTCG+GAAGCCTA#0/1, got 
A00204:533:HC5CLDSXY:1:1101:10203:1000:N:0:CTAGTTCG+GAAGCCTA#0/2

However I don’t think we shall expect an id ends with #0/1, which is a R1, in a *.2.fastq (R2). I have another linux machine with bmtagger downloaded from the same place but with kneaddata v0.7.2 and it processes the exact files fine. I was wondering what could have gone wrong.

Any help would be appreciated.

Best,
Huan

1 Like

Having the same issue. I am working with reads from NCBI, and renamed the reads with the /1 and /2 suffixes and removed spaces as suggested in this post: Is it necessary to run trimmomatic before HUMAnN? My read number decreased from around 10000000 to 7000000 in forward read and 2000000 in reverse read but still getting this error.

Hi @fanhuan @michaelwoodworth, were you able to solve the issue?
I am having the same issue.