TopHat color space

[SongLi]

Hi All,

I have trouble using tophat with files downloaded from SRA.

My command is:

./tophat -C -p 5 -o ./825tophat ../bowtie-0.12.7/indexes/ath_gmc_colspace_110510 ~/SRR039825.fastq

Error encountered parsing file /home/SRR039825.fastq:
Length mismatch between sequence and quality strings for SRR039825.1 923_6_55 (36 vs 36).

The sequence is here:
@SRR039825.1 923_6_55
T00310202021210203103230203233012210
+
!;>1<998495<<3$4.40%/87-101*&3,8#%'#

I dig into the code, and find the problem is at line 931-934 of tophat.py, where the length of the sequence has to be 1 character longer than the quality score.

Why is this and how can I fix it?

Thanks,

Song Li

[SongLi]

A little update on this issue:

I wrote a script that trim off the first quality value in my fastq file. Then tophat runs smoothly through the whole analysis.

I am still not sure that's the correct way of solving this problem.

Thanks,

Hi All,

I have trouble using tophat with files downloaded from SRA.

My command is:

./tophat -C -p 5 -o ./825tophat ../bowtie-0.12.7/indexes/ath_gmc_colspace_110510 ~/SRR039825.fastq

Error encountered parsing file /home/SRR039825.fastq:
Length mismatch between sequence and quality strings for SRR039825.1 923_6_55 (36 vs 36).

The sequence is here:
@SRR039825.1 923_6_55
T00310202021210203103230203233012210
+
!;>1<998495<<3$4.40%/87-101*&3,8#%'#

I dig into the code, and find the problem is at line 931-934 of tophat.py, where the length of the sequence has to be 1 character longer than the quality score.

Why is this and how can I fix it?

Thanks,

Song Li

[xinwu]

A little update on this issue:

I wrote a script that trim off the first quality value in my fastq file. Then tophat runs smoothly through the whole analysis.

I am still not sure that's the correct way of solving this problem.

Thanks,

This is due to the format used by NCBI. NCBI transforms all the data from different platforms to a standard FASTQ format.
Tophat uses bowtie for reads mapping and it expects csfasta and qual files if the data is color-spaced. Sequence in csfasta has additional 'T' adapter comparing to qual file, so tophat expects one more base. Just tell bowtie you use fastq format rather than fasta.

[ngsbioinfo]

hi all,

I am new to NGS analysis field. I am working on RNA-Seq data, aim it to identify all novel junctions and transcripts. I would appreciate if any one can help me out in using tophat and cufflinks for that matter.

Thanks in advance