Kallisto quant error "std::bad_alloc"
0
0
Entering edit mode
6 months ago
  • Windows 10
  • Running in Terminal

I'm working on a non-model organism. I concatenated two samples raw reads (larval and adult), and used Galaxy Trinity to assemble the concatenated forward and reverse strands.

I then ran kallisto index on the fasta file (~107,500 KB)

kallisto index -i I.f_concat_Trinity_assembled.index I.f_concat_Trinity_assembled.fasta

This generates a file (size 1,186,659 KB) with the following log

[build] loading fasta file I.fasciola_109A1_67B8_concat_Trinity_assembled.fasta
[build] k-mer length: 31
[build] warning: clipped off poly-A tail (longer than 10)
        from 78 target sequences
[build] counting k-mers ... done.
[build] building target de Bruijn graph ...  done 
[build] creating equivalence classes ...  done
[build] target de Bruijn graph has 512211 contigs and contains 54818622 k-mers

When I run kallisto quant using the index file

kallisto quant \
-i I.f_concat_Trinity_assembled.index \
-o 67B6 \
-t 4 \
67B6_1.fq.gz \
67B6_2.fq.gz \
&> 67B6.log

I get the following log and error message

[quant] fragment length distribution will be estimated from the data
[index] k-mer length: 31
[index] number of targets: 175,677
[index] number of k-mers: 54,818,622
[index] number of equivalence classes: 65,536
terminate called after throwing an instance of 'std::bad_alloc'
  what():  std::bad_alloc

I have no idea how to fix this problem. I have run kallisto using practise human files and it worked so I can't quite work out what's going wrong. Googling tells me I may not have enough memory, but I have over 100 GB storage free, and 16 GB RAM. The human index I generated and aligned to was 2,445,321 KB (over twice the size), so really size shouldn't be a problem.

What is std::bad_alloc? How do I fix it?

RNA-Seq kallisto quant • 325 views
ADD COMMENT
0
Entering edit mode

Fixed it - switched computer off for 20 minutes, turned it back on, reran the kallisto index on the original FASTA to generate a new index, and ran quant against then new index - worked flawlessly.

To quote the IT crowd

Have you tried turning it off and on again?

ADD REPLY
0
Entering edit mode

A similar problem has been discussed in Github.

ADD REPLY

Login before adding your answer.

Traffic: 1274 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6