martindemko
commented
5 years ago Recently I found out, that changing parameter -word_size from 12 to any higher number will do the magic and save the analysis.
Open martindemko opened 5 years ago
martindemko
commented
5 years ago Recently I found out, that changing parameter -word_size from 12 to any higher number will do the magic and save the analysis.
Hi, I got an error on my dataset which says:
hs-blastn: sources/interval_tree.cpp:487: Boolean IntervalTree::IntervalTreeContainsHSP(const HSP*, Int4): Assertion `hsp->s_off >= s_min' failed. Aborted (core dumped)
Command:
hs-blastn align -db mirna_contaminants.fasta -window_masker_db mirna_contaminants.fasta.counts.obinary -query PaC115.fasta -reward 1 -penalty -3 -gapopen 0 -gapextend 2 -word_size 12 -max_target_seqs 1 -num_threads 10 -outfmt 6 -out PaC115.tsv
However, I really don't understand why did this happen. I tried it several times and everytime I got the exact same output (incomplete at the same line and character). I tried to remove that particular sequence (which is the one on the last broken line) and it happened again but for another sequence. I tried it even run just for that particular sequence and everything went well. I also tried to parse my fasta input into many fasta files with only one sequence and no error. So I'm out of ideas now.
Could you help me somehow, please?
PS: the project is about mapping miRNA sequences to our rather short DB of small RNA seqs (i.e., contaminants). They are very short (16-28 bp).