kallisto index problem

Question

[anagonzangel]

Hi all

I am trying to create an index for kallisto:

kallisto index -i compiled_A.idx --make-unique cladeA_Rosic.fasta microadriaticumTx.fasta

Kallisto starts running but shows 0 contigs, and 0 kmers. It creates an output file but it is an empty file. The files I am using to make the reference are in fasta format, and look ok to me:

>gi|659488139|gb|GBGW00000000.1|GBGW01000000 TSA: Symbiodinium sp. A2, transcriptome shotgun assembly

>gi|659488138|gb|GBGW01000001.1| TSA: Symbiodinium sp. A2 Locus_1219_Transcript_1/2_Confidence_1.000_Length_4042 transcribed RNA sequence
GGCCGAAATTGGCGGCATCAGGCTCGTAAGCAGCGACGGCAAATGCATCAGCTGCACGACCCCATTGGTG
TGCAGCACTGCAGTTGCGGTAGTGAAGGGGCCCGTGGCGAAAGCCTGCAATCCCAACCCCCAGCCGGGGG
ACATGTTTGGTTGGAAGGTGAACAAGAGCTCACCATACGATATTCGGGTAGACTTCCCGCTGGAGAAGCT

and

>gi|556422487|gb|GAKY00000000.1|GAKY01000000 TSA: Symbiodinium sp. A1, transcriptome shotgun assembly

>gi|556422486|gb|GAKY01000001.1| TSA: Symbiodinium sp. A1 Locus_1_Transcript_1/4_Confidence_0.444 transcribed RNA sequence
CGCTGTTGGAGTTGAAGCTTTTGCCACCCGATGTTGTGCAAATGACCTGCTACTGCTAAAGCAGCTGCAG
CTCCCGCTTTGCCAGCAATGGCCAGTCCGGACCGCCAAGATCGATTGGGTCAGGCGGTCGCTCTGGAGAA
GAAGGCACGTGAGCTCGAAGGACAGGGCCAATTACAGGATGCCCTTCGCGAGTACCAAAACTGCTTGCAG

I don't know why this is not working. Has anyone ever encountered this before?

Thank you!

[GenoMax]

Looks like you have sequences with 0 bases. Perhaps that is the problem.

Answer 1

[anagonzangel]

Thank you for your observation. I have an update on this. Kallisto works when you get rid of empty sequences. To get rid of them I used fastasort (to sort) and nano (to manually delete the sequences).

[igor]

To do it automatically, see this previous discussion: Removing All Empty Fasta Sequences From A File (Was: Editing The Headers Of The Fasta Format Sequence)