Stringtie/Ballgown input files

[Vince Schulz]

I am having difficulties getting Ballgown to read data output from Stringtie.

I map reads with tophat2, find transcripts with stringtie, merge the transcripts on different samples with cuffmerge, and then generate count data with stringtie like
stringtie accepted_hits.bam -o ./stringtie2.gtf -p 4 -G merged.gtf -B -e

This generates the same stringtie2.gtf for each sample, but the i_data.ctab files are different in that some lines are missing from one compared to another, eg

i_id    chr    strand    start    end    rcount    ucount    mrcount
1    chr1    +    12228    12612    0    0    0.00
2    chr1    +    12722    13220    1    0    0.33
3    chr1    -    14830    14969    83    0    26.92
4    chr1    -    15039    15795    22    0    6.78
5    chr1    -    15948    16606    25    0    9.17
6    chr1    -    16766    16857    25    0    12.50

see line 5  vs below for another sample

i_id    chr    strand    start    end    rcount    ucount    mrcount
1    chr1    +    12228    12612    0    0    0.00
2    chr1    +    12722    13220    0    0    0.00
3    chr1    -    14830    14969    155    0    44.92
4    chr1    -    15039    15795    45    0    10.18
5    chr1    -    16766    16857    204    0    98.65


Ballgown complains about the input:
Error in ballgown(samples = c(...  :
  intron ids were either not the same or not in the same order across samples. double check i_data.ctab for each sample.

I can try using Tablemaker  per the Ballgown vignette, but the stringtie manual suggests that I should be able to use stringtie output directly.

I am using stringtie-1.0.4.Linux_x86_64 and ballgown_2.0.0
R version 3.2.0 (2015-04-16)

Thanks,

Vince

[Vince Schulz]

Sorry about the previous post,  it looks like I used an incorrect GTF file for one of the samples.