TopHat 1.1 failing on colorspace SE reads

[krobison]

I'm trying to analyze a single end dataset from SRA with the brand new version of TopHat (1.1). TopHat crashes with the below error message, and looking at this & the code it appears that even with single ends it is trying to run a validation check on the 2nd set of reads

  File "/usr/local/bin/tophat", line 2093, in main
    params.read_params = check_reads(params.read_params, left_reads_list + "," + right_reads_list)
TypeError: cannot concatenate 'str' and 'NoneType' objects

If I replace line 2093 with

        if params.skip_check_reads == False:
            if right_reads_list !=None:
                params.read_params = check_reads(params.read_params, left_reads_list + "," + right_reads_list)
            else:
                params.read_params = check_reads(params.read_params, left_reads_list)

Then I get farther but hit a new error

Length mismatch between sequence and quality strings for SRR040290.1 VAB_ugc_85__100_137__138_121__123_bc_Frag50_solid0032_20090715_ugc_121__1231_49_36 length=50 (51 vs 51).

I'm too worn out to puzzle how to get past that one -- my best guess is this is related to the "extra" colorspace value which bowtie option "--col-keepends" deals with

[kopi-o]

Yep, I just got the same error message (the first one; haven't tried to modify the code). I'm also using single-end color space reads (.csfasta + .qual)

[Telor]

Yes, Just to confirm that you are not the only one, I'm getting this error too, but on standard single end reads (not color-space).

[sunnyjoy]

Google search the same error message leads me here. Same problem with single-end color space reads.

[DerSeb]

same here.

[Daehwan]

Hi guys,

I'm Daehwan, who made this bug and fixed it, you can grab a fixed version at http://tophat.cbcb.umd.edu/index.html

Thanks

[krobison]

Thanks for the rapid fix!!

[DerSeb]

great, i will try it right now! thx for the great support!

[Telor]

Great. Thanks for the speedy fix. It looks like its working fine so far (fingers crossed )

[DerSeb]

Hello. I can now successfully start and get passed the first error encountered here. However, I still run into the next error mentioned above:

Tue Oct  5 16:10:32 2010] Beginning TopHat run (v1.1.0)
-----------------------------------------------
[Tue Oct  5 16:10:32 2010] Preparing output location /home/schaefer/tophat/RBM20/Sample14//
[Tue Oct  5 16:10:32 2010] Checking for Bowtie index files
[Tue Oct  5 16:10:32 2010] Checking for reference FASTA file
[Tue Oct  5 16:10:32 2010] Checking for Bowtie
	Bowtie version:			 0.12.3.0
[Tue Oct  5 16:10:32 2010] Checking for Samtools
	Samtools version:		 0.1.8.0
[Tue Oct  5 16:10:39 2010] Checking reads

Error encountered parsing file ...fastq:
 Length mismatch between sequence and quality strings for 853_8_25/1 (49 vs 49).

When I check the fastq file, everything seems fine:

@853_8_25/1
GNNGTGNTNCANNCGTNNGAGNNCACNNACANCCGANNACGNAAAGNAN
+
*""%%%"%"%%""%%%""%)&""%%%""%%+"&'%&""'(%"'))'"&"
@853_8_35/1
CNNACGNANACNNACCNNCCGNNTAANNNNGNGAACNNCNANCNCNNTN
+
:""=54"@"=+""A98""745"";98""""2"=@>8""<"4"<"6"";"
@853_8_75/1
GNNACCNCNTCNNAACNNTACNNCGANNGTGNGGACNNGTCNCGAGNCN
+
="";<7"0";4""=;:"">;=""94<"".,5".;26""9%)"(%(("("
@853_8_96/1
...

Is this error still occuring to someone else?

[Daehwan]

DerSeb, what's your command?

[DerSeb]

This is my command:

tophat -G /data/genetics/datasets/genome-annotation/ensembl-56/Rattus_norvegicus.RGSC3.4.56.gtf -o /home/schaefer/tophat/Sample14 -C rn4_c /home/schaefer/tophat/fastq/Sample_14_Qual.fastq

[Daehwan]

Since you are using -C, which is for colorspace read, you need to use colorspace reads instead of nucleotide reads.

[DerSeb]

I see... I converted my CS reads to fastq, using scripts supplied with MAQ (solid2fastq.pl or fq_all2std.pl csfa2std). They convert the cs values to letters mimicking a "pseudo" genetic sequence.

I will look into this and see how I can change this.

Thx!

[DerSeb]

I have now started a thread dedicated to reformatting SOLiD reads for TopHat:

http://seqanswers.com/forums/showthread.php?p=26692