y9c / cpup

Convert mpileup format to base count tsv table

cpup

Convert samtools mpileup result into base count table

Feature:

• multiple bam files are supported
• filter sites before output
• output by strandness
• group and count indels by pattern

Install

git clone https://github.com/y9c/cpup.git
cd ./cpup
make

Usage

Pipe the output of samtools mpileup into this tools!

Usage:
  samtools mpileup -d 0 -Q 10 --reverse-del -l <.bed> -f <.fa> <.bam> | cpup

  -h, --help          show help
  -H, --headerless    hide header
  -S, --strandless    ignore strand information
  -s, --by-strand     output by strand
  -m, --major-strand  output major strand only
  -i, --indel         append indel count
  -e, --ends          append read ends (5' | 3') count
  -c, --count []      select count columns
  -f, --filter []     filter sites
  -F, --drop []       drop sites

samtools mpileup can mpileup the mapping result site by site in the format below. The 5th (8th, 11st, 14th, ...) column report the observed bases in each site.

XII     455422  C       16      <<<<<<<<<<<<<<,,        FFFFFFFFFFFFFFFF        4       <<,,    FFFF
XII     455423  T       16      <<<<<<<<<<<<<<,,        FFFFFFFFFFFFFFFF        4       <<,,    FFFF
XII     455424  C       17      <<<<<<<<<<<<<<,,^$,     FFFFFFFFFFFFFFFFE       4       <<,,    FFFF
XII     455425  A       18      <<<<<<<<<<<<<<,,,^$,    FFFFFFFFFFFFFFFFFF      4       <<,,    FFFF
XII     455426  A       18      <<<<<<<<<<<<<<,,,,      FFFFFFFFFFFFFFFFFF      4       <<,,    FFFF
XII     455427  A       20      <<<<<<<<<<<<<<,,,,^$,^$,        FFFFFFFFFFFFFFFFFFEE    4       <<,,    FFFF
XII     455428  C       20      <<<<<<<<<<<<<<,,,,,,    FFFFFFFFFFFFFFFFFFFF    4       <<,,    FFFF
XII     455429  A       20      <<<<<<<<<<<<<<,,,,,,    FFFFFFFFFFFFFFFFFFFF    4       <<,,    FFFF
XII     455430  G       22      <<<<<<<<<<<<<<,,,,,,^$,^$,      FFFFFFFFFFFFFFFFFFFFBB  4       <<,,    FFFF
XII     455431  G       23      <<<<<<<<<<<<<<,,,,,,,,^$,       FFFFFFFFFFFFFFFFFFFFEED 4       <<,,    FFFF
...

cpup can convert the mpileup output into table below by parameters -i -s -f mut:3.

chr     pos     ref_base        strand  depth,a,c,g,t,n,gap,insert,delete,istat,dstat           depth,a,c,g,t,n,gap,insert,delete,istat,dstat
XII     455496  C       -       1409,17,0,0,0,0,2,0,2,,c:1|caa:1        139,3,0,0,0,0,1,0,2,,c:1|caa:1
XII     455498  T       -       1454,0,6,2,0,0,407,0,244,,at:220|att:24 144,0,0,1,0,0,37,0,27,,at:23|att:4
XII     455499  T       -       1496,0,1,12,0,0,250,0,12,,t:7|tt:5      151,0,0,2,0,0,29,0,1,,t:1
XII     455500  A       -       1455,0,2,2,0,0,256,0,0,,        145,0,1,0,0,0,28,0,0,,
XII     729179  T       +       223,0,3,0,0,0,33,0,0,,  56,0,0,0,0,0,14,0,0,,
XII     729182  C       +       235,1,0,0,4,0,2,0,0,,   56,0,0,0,0,0,1,0,0,,

NOTE:

• -Q in samtools mpileup should not be set to zero, which might bring bug in counting overlapping reads.
• -f to check any (max) value greater than cutoff. eg: mut:3 for filter sites with more than or equal to(>=) 3 mutations in any sample. You can set multile filters, like mut:3,delete:2 to filter sites with >= 3 mutations in any samples, meanwhile, are should be >=2 delete events.
• -F to check all (min) value greater than cutoff.

Q&A?

• filter input base by its quality?

samtools can do it

• filter output site by cutoff?

awk can do it

• samtools mpileup is too slow?

bcftools mpileup might a better choice, if you are not confusing with tags (-t AD, -t ADF, -t ADR).

cpup's Issues

ref_base not equal between samples

Hi cpup,
I am using your tool like in the below command and noticed that the ref_base values are not always the same between samples, despite between aligned to the same reference and passing the same genome/annotation. Is there an explanation for this?

samtools mpileup -aa -l annots.bed -f genome.fa my.bam | cpup --strandless --ends --indel > out.txt

Thanks,
Joe