Errors in processing RNA-Seq data

[Steve Piccolo]

I'm getting an error when trying to map my RNA-Seq data to the human genome. Below is an example of the errors I'm getting. Below that, I've included the command-line arguments I'm using. And below that I've included the top part of my FASTQ file. Any ideas? Thanks in advance.

[-/15] Matching 18437696 sequences of: /fslhome/myhome/SRR002322.fastq

Reads per processor: 128

ERROR at 5,1.

        Value is: a(97)

ERROR at 5,1.

        Value is: v(118)

ERROR at 5,1.

        Value is: J(74)

ERROR at 3,7.

        Value is: 3(51)

ERROR at 4,8.

        Value is: ^@(0)

ERROR at 1,5.

        Value is: a(97)

ERROR at 4,8.

        Value is: ^@(0)

ERROR at 1,5.

        Value is: ^@(0)

ERROR at 6,10.

        Value is: ^@(0)

[15/0] 0% reads complete

ERROR at 5,1.

        Value is: ^@(0)

[2/0] 0% reads complete

[10/0] 0% reads complete

[12/0] 0% reads complete

[9/0] 0% reads complete

[13/0] 0% reads complete

[6/0] 0% reads complete

[3/0] 0% reads complete

[7/0] 0% reads complete

[8/0] 0% reads complete

[0/0] 0% reads complete

[11/0] 0% reads complete

[1/0] 0% reads complete

[4/0] 0% reads complete

[14/0] 0% reads complete

[5/0] 0% reads complete

ERROR at 5,1.

        Value is: ^@(0)

ERROR at 3,7.

        Value is: 5(53)

ERROR at 6,2.

        Value is: ?(-80)

ERROR at 6,2.

        Value is: t(116)

ERROR at 5,9.

        Value is: ^@(0)

ERROR at 3,7.

        Value is: t(116)

ERROR at 8,4.

        Value is: ?(-85)

ERROR at 5,9.

------------------------------------------------------------------------------------------------------------

PROG=~/gnumap/bin/gnumap

GENOME="$(ls ~/hg19/chr*.fa)"

SEQFILE=~/SRR002322.fastq

OUTFILE=$SEQFILE.out

/usr/mpi/fsl_openmpi_gcc-1.4.2/bin/mpiexec -np $nproc -machinefile $machfile \

   $PROG -g \"$(echo $GENOME | sed -e 's/ /,/g')\" -o $OUTFILE \

   -v 1 -m 10 -j 5 -c 8 \

   --MPI_largemem \

   $SEQFILE > KNELL.out

------------------------------------------------------------------------------------------------------------

@SRR002323.1 080317_CM-KID-LIV-2-REPEAT_0003:3:1:112:566

TGGGGTTGTGATTTTGATATTGGTGGATTGAGGGTT

+

IIIIIIIIIIIIIIIII=&IIIIIIIIAIIIIIII:

@SRR002323.2 080317_CM-KID-LIV-2-REPEAT_0003:3:1:121:542

TATGAATATGCAAGAAGGCATCCTGATTACTCTGTC

+

IIIIIIIIIIIIIIIIIIIII0IIIIIIIIIIIII6

@SRR002323.3 080317_CM-KID-LIV-2-REPEAT_0003:3:1:123:599

TCTGTATCTGGTCCTGTGTTACTGTAGTGGTAATTA

-------------------------------------------------------------------------------------------------------------

[Steve Piccolo]

I sent this email out a while again and am now just coming back to it. So far I haven't received any replies. But I also noticed that I'm getting the following error message in the standard out:

WARNING:  --illumina flag caused undesirable side-effects.

        Turning it off

Does that ring any bells on what might be going on here?

Thanks,

-Steve

[Nathan Clement]

What version of GNUMAP are you using?

Nathan

[Steve Piccolo]

Hi Nathan,

I just updated to the latest version and ran it again. I'm still seeing the following message:

WARNING:  --illumina flag caused undesirable side-effects.

    Turning it off

I don't get this message with some other FASTQ files, so there must be something specific to these FASTQ files that is causing the error. At the bottom of this message is a sample of the problematic FASTQ file.

Thanks,

-Steve

[Nathan Clement]

Steve,

I'm guessing it's a problem with one of the sequences in your fastq file.  Sometimes there are weird characters in it that cause the program to fail later in the mapping process.  If you can narrow it down to just a couple of sequences (or at least a file that would be reasonable to send via email), that would be ideal.  If you can't narrow it down at all, or if it actually is a bigger problem with the program, I might need to look at your specific fastq file and genome to isolate the root problem

Nathan