divide by zero in adaptive seeding

[h-2]

via email:

I am having problems with running lambda on part of my data and hope you can help if possible.

I am trying to align some NGS reads that I've processed into 50bp kmers to a custom index and while it successfully aligned the first 9 samples and outputted normally, I have 6 other samples that exit lambda with a Floating point exception and I don't understand why this is happening:

> lambda -q IN-M1O2_S10_L001_R1_001.fastqKMER.fastq -i INM1O2.fasta.lambda -o kmerS10R1.m8 -p blastn


LAMBDA - the Local Aligner for Massive Biological DatA
======================================================
Version 1.9.0

Loading Subj Sequences... done.
Loading Subj Ids... done.
Loading Database Index... done.
Loading Query Sequences and Ids... generating reverse complements... done.
Searching and extending hits on-line...progress:
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

|Floating point exception

I have found this error is reproducible on different computers running mint and ubuntu respectively (both are i7s with 16Gb and 64Gb ram) and occurs using both Lambda version 1.9.0 and version 1.9.1.

[h-2]

fixed by a842592