TopHat Fusion post (Error)

[samar]

Hello,
I am trying to run TopHat Fusion Post, and I am having this error:
Traceback (most recent call last):
File "/N/u/lab/tophat-2.0.8b.Linux_x86_64/tophat-fusion-post", line 2627, in ?
sys.exit(main())
File "/N/u/lab/tophat-2.0.8b.Linux_x86_64/tophat-fusion-post", line 2595, in main
map_fusion_kmer(bwt_idx_prefix, params, sample_updated)
File "/N/u/lab/tophat-2.0.8b.Linux_x86_64/tophat-fusion-post", line 317, in map_fusion_kmer
subprocess.call(cmd, stdout=open(output_dir + 'fusion_seq.bwtout', 'w'), stderr=open('/dev/null', 'w'))
File "/usr/lib64/python2.4/subprocess.py", line 419, in call
return Popen(*args, **kwargs).wait()
File "/usr/lib64/python2.4/subprocess.py", line 550, in __init__
errread, errwrite)
File "/usr/lib64/python2.4/subprocess.py", line 996, in _execute_child
raise child_exception
OSError: [Errno 2] No such file or directory

Itried with different solutions, such as using a new directory for Bowtie both 1 and 2. Make the directory similar to what it is suggested in thier website.

Can anyone please help me with suggestions.
Thank you,
Samar

[Jon_Keats]

First follow their folder structure exactly.

Second, in my experience and frustration you need to use their hg19 bowtie1 reference index

Third, depending on your samples (read length & number of reads) fusion-post will run out of memory. In one of our compute farms we spec, 80 million 50x50 paired reads will work > 90% of the time. Over 100 million pairs fails almost constantly. This is with 24 Gb or Ram available to the job. On trick is not to max out the parallel computing so only use 1/2 the available processors but take all the memory.

Fourth, if you have not run the test samples try that to ensure you have blast set correctly as that can be a nightmare. You can inactive the blast step to test but it really helps remove most of the false results.

[samar]

First follow their folder structure exactly.

Second, in my experience and frustration you need to use their hg19 bowtie1 reference index

Third, depending on your samples (read length & number of reads) fusion-post will run out of memory. In one of our compute farms we spec, 80 million 50x50 paired reads will work > 90% of the time. Over 100 million pairs fails almost constantly. This is with 24 Gb or Ram available to the job. On trick is not to max out the parallel computing so only use 1/2 the available processors but take all the memory.

Fourth, if you have not run the test samples try that to ensure you have blast set correctly as that can be a nightmare. You can inactive the blast step to test but it really helps remove most of the false results.

Jon_Keats, thank you very much for replying!
Yes. I tried to follow their exact directory format.
And for the Bowtie, I used Bowie two because I think TopHat 2 is using this index.
I tried for the post fusion step with their Bowtie 1 and 2 but it did not work.
In the current time, I am not using the blast because I am trying to make things work. Hopefully.
Is it better to try to run everything with their samples? Or did you mean with any other sample?

Thank you,
Samar

[samar]

Issue solved..
was not related to tophat..from PUTTY
Thanks

[jp.]

Hi there
I have used tophat fusion with hg19 without problem. But now with UCSCmm10, I got error at fusion post:
BLAST Database error: No alias or index file found for nucleotide database [blast/nt] in search path

I have downloaded blast:
1. ncbi-blast-2.2.28+
2. extracted est_mouse.tar.gz mouse_genomic_transcript.tar.gz within ncbi-blast-2.2.28+/ and ncbi-blast-2.2.28+/bin

I am not sure about other_genomic* and nt* ?
I used and exported PATH=$PATH:bowtie1, tophat2, blast, samtools
Can some one help me ?
Thanks

[jp.]

Issue solved..
was not related to tophat..from PUTTY
Thanks

Can you explain how did you solve error in blast
I have downloaded blast and all database and kept in NCBI-blast****/db

What is next to do with it ??
thank you