problems of sample names for combine_mpa.py

[chunchenghuang]

Hi,
Thank you for your script, your script has been extremely helpful. However I was trying to use kreport2mpa.py with --display-header for 63 samples, combine_mpa.py, which is said to be able to use the header automatically in the manual reported errors:
Number of files to parse: 63

dir="."
for report in $(find $dir -maxdepth 1 -name "*_report.txt"|sort -V);do
report=${report//_report.txt/};
#ktImportTaxonomy -m 3 -t 5 ${report}_report.txt -o ${report}_krona.html
kreport2mpa.py --display-header --no-intermediate-ranks -r ${report}_report.txt -o ${report}_mpa.txt
done
combine_mpa.py -i *_report_mpa.txt -o combined_mpa.tmp

Traceback (most recent call last):
  File "/opt/KrakenTools/combine_mpa.py", line 142, in <module>
    main()
  File "/opt/KrakenTools/combine_mpa.py", line 83, in main
    [classification, val] = line.strip().split('\t')
ValueError: too many values to unpack

This is my bash script for running the thing (I tried to sort my files by sort -V in bash to replace the header by names or the sample input order manually without the --display-header, commands but the output order seems to be different anyway)

dir="."
for report in $(find $dir -maxdepth 1 -name "*_report.txt"|sort -V);do
    report=${report//_report.txt/};
    #ktImportTaxonomy -m 3 -t 5 ${report}_report.txt -o ${report}_krona.html
    kreport2mpa.py --no-intermediate-ranks -r ${report}_report.txt -o ${report}_mpa.txt
done

file_list=$(echo *_report.txt|tr " " "\n" | sort -V |tr "\n" " " )
echo *_report.txt | tr " " "\n" | cut -d "_" -f1 | sort -V > sample.header
echo "#Classification" > tmp.tmp
cat tmp.tmp sample.header | tr "\n" "\t" > table.header
combine_mpa.py -i $file_list -o combined_mpa.tmp
sed 1d combined_mpa.txt > table.tmp
cat table.header table.tmp > combined_mpa.txt
rm *.header
rm *.tmp

[mars188]

were you able to find a solution?