end of file unexpected error when running splicemap

[kanewong]

Hi all,

Nice to meet you, I am a newbie to this forum. Hopefully I am making post in the right forum.

I am running splicemap to find splice-junction of a set of RNA-seq data but got the following error message:
.
.
.
.
.
Total Bowtie mapping section execution time: 5533.83 s.
Generating .t file!...
Total .t creation section execution time: 470.855 s.
Removing mapping results!...
Creating mapping index!...
Generating mapping file!...
Total mapping index creation section execution time: 445.987 s.
Total Mapping section time: 6451.07 s.
Running SpliceMap on gi|240254421|ref|NC_003070.9| ...
sh: 1: Syntax error: end of file unexpected
Running SpliceMap on gi|240254678|ref|NC_003071.7| ...
sh: 1: Syntax error: end of file unexpected
Running SpliceMap on gi|240255695|ref|NC_003074.8| ...
sh: 1: Syntax error: end of file unexpected
Running SpliceMap on gi|240256243|ref|NC_003075.7| ...
sh: 1: Syntax error: end of file unexpected
Running SpliceMap on gi|240256493|ref|NC_003076.8| ...
sh: 1: Syntax error: end of file unexpected
Total SpliceMap section time: 35.1471 s.
Merging results and computing junctions...
Reading SAM headers
ERROR: could not open temp/NC_003070.fna_84.sam
ERROR: Amalgamation failed = 512

Does anyone know what it means? I tried running the same set of data using tophat but no error message was produced.
Thank you for your help

[kanewong]

Here is part of the original script(run.cfg) provided by the developers of spliceMap,

################################################################
# These are the two lists of sequencer reads files.
# "reads_list2" can be commeted out if reads are not paired-end.
# Make sure the order of both lists are the same!
# Also, "reads_list1" must be the first pair.
# Note: pair-reads should be in the "forward-reverse" format.
# (multiple values)

> reads_list1
data/long_reads_1_100K.txt.seq
<

> reads_list2
data/long_reads_2_100K.txt.seq
<
###################################################################

Then I change the command according to my real situation
###################################################################


> reads_list1
data/SRR360147_1.fastq
<

> reads_list2
data/SRR360147_2.fastq
<
###################################################################


My dataset is here: http://www.ncbi.nlm.nih.gov/sra/SRX103665

[GenoMax]

Here is part of the original script provided by the developers of spliceMap,


###################################################################
Then I change the command according to my real situation

> reads_list1
data/SRR360147.fastq
<

#>
#<
###################################################################

Try to use "#" as shown above to comment out the read2 list. See example in red above. I am not completely sure what script you are referring to but see if that helps.

[kanewong]

(del).......

[GenoMax]

Hi, thanks for your advice, finally I download the same data from ebi(different file format for pair end reads), and modify my script as follows:


###################################################################
> reads_list1
data/SRR360147_1.fastq
<

> reads_list2
data/SRR360147_2.fastq
<
###################################################################

Unfortunately an error still produced

Was the error same as the one you originally posted or different?

[kanewong]

Was the error same as the one you originally posted or different?

same..........

[GenoMax]

Are there files in "temp/*" directory that the program is trying to access? Has the bowtie alignment worked?

[kanewong]

The temp directory is produced after I start running the program, but I am not sure if the program is accessing the files in temp when the error occurs.

Bowtie alignment takes place before the error message appears, so I guess it works. But I am not sure if there is still any bowtie alignment steps afterwards.

I learn that some programmers check the error logs to figure out which file produces error, but I am not sure how to do that.

[GenoMax]

Here is a sample page of how one can capture the standard error and output to file(s). The instructions are shell specific so depending on what shell (C-family or Bash) you are using you will need to choose the right directions.

If you can include the entire command line (along with the error logs or the relevant parts with errors) that would help.