Kallisto is freezed with some samples..

[Antonio Rodriguez Franco]

I have 24 files corresponding to 12 paired-end fastq reads

I have mapped all of them with Kallisto, and it takes around 3 min in my computer for the mapping to be accomplished for every pair of paired-end files

But there are two files corresponding to paired reads that are not working. Kallisto is working for more than 24h and freezes in the step of finding pseudoalignments

kallisto quant -i Mus_musculus.index -o Kallisto/A2 -b 100 -t 7 A2_1.fastq.gz A2_2.fastq.gz 

[quant] fragment length distribution will be estimated from the data

[index] k-mer length: 31

[index] number of targets: 61,440

[index] number of k-mers: 34,582,631

[index] number of equivalence classes: 128,766

[quant] running in paired-end mode

[quant] will process pair 1: A2_1.fastq.gz

                             A2_2.fastq.gz

[quant] finding pseudoalignments for the reads ... (24h working without success)

I have checked that the number of reads included in every file is the same, so does the head and the tail of these files indicating that they contain synchronized reads withour orphans

Any clue or hint about this ?

[Pall Melsted]

Hi Antonio

Do the files look alright with seqtk or similar programs, i.e. they are not corrupted?

Can you run this again (I'm assuming you've tried) or on a different machine (if it doesn't finish within 15 minutes, you can probably kill it)?

If the data can be shared I'd be happy to take a look at what is happening and debug the issue. 

Best regards,

Pall

--
You received this message because you are subscribed to the Google Groups "kallisto-sleuth users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to kallisto-sleuth-users+unsub...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/kallisto-sleuth-users/8c3ca392-485f-43a6-9560-471914184864%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Antonio Rodriguez Franco]

From what I can tell you, FastQC can analyze both fastq files seamless.

A counting of the number of lanes with a wc -l does not give any problem

So I believe that the integrity of the files are correct

I can share these files through DroxBox or sent you by Wetransfer

[Pall Melsted]

Great. I'll follow up off-list.

- Pall

To unsubscribe from this group and stop receiving emails from it, send an email to kallisto-sleuth-users+unsubscri...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/kallisto-sleuth-users/8c3ca392-485f-43a6-9560-471914184864%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "kallisto-sleuth users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to kallisto-sleuth-users+unsub...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/kallisto-sleuth-users/a8524c18-882c-4dbf-854f-b95aa69d31e2%40googlegroups.com.

[Matus Medo]

Dear Pall,

The same thing happens to me - Kallisto hangs at "[quant] finding pseudoalignments for the reads ..." for one file of many that I have. May I ask if you have managed to find out the reason behind Antonio's identical problem?

Best regards,

matus

[Matus Medo]

One thing occurred to me and it helped me to solve the problem - I was analyzing the files on another computer and upon transfer, something went wrong and one of the files got corrupt, so it seems. I copied the file again and pseudoalignment went fine now. It would be nice if Kallisto would not hang up like this and report that there is a problem with the input file but I am not sure if it is really worth looking into that (now). In any case, thanks for this nice software!

Best,

matus