blastn error values meaning

Question

[sebastien.letort]

Hi,

my blast command ends with an error value 9 and I can't find the meaning of it. Can someone give me a clue on what went wrong ?

blaster -n blastn -q Boleracea_db/batches/batch_244.fa -s Boleracea_db/Boleracea_chunks.fa -B debug_batch_244.fa -a -S 0 -E 1e-300 -L 100 -I 90 -N -p '-a 1 ' -c -v -3
Beginning Blaster (version 2.25)
mar. févr. 20 16:56:09 UTC 2018
Exception catch:  program blastn return with error value: 9, stopping blaster
Exit with value 1

[lieven.sterck]

In the context of the REPET Te-detection pipeline , I assume?

[sebastien.letort]

Yes (how do you know ?) but the error occur also if I run the blaster command alone. I blast chunks of the fasta file to seek for the bad sequence.

[lieven.sterck]

haha :-) , REPET user myself and I recognized the syntax (as well as the blaster)

Do you also get the error if you only run the blastn cmd? (so not in the blaster context)?

[sebastien.letort]

So, I isolated the sequence that pose bp (at least one sequence). https://data-access.cesgo.org/index.php/s/drp18eXmrj1a5lL

I ran blaster on it -> error value 9 I ran blastn => no error. I'm not used to blaster and not really to blast, so I used the following option conversion : blaster

-n blastn                     blastn
-q s1.1.fa                    -query s1.1.fa
-s Boleracea_chunks.fa        -db Boleracea_chunks.fa_cut 
-B debug_err_s1.1             -out debug_err_s1.1
-a                            ?
-S 0                          ?
-E 1e-300                     -evalue 1e-300
-L 100                        ?
-I 90                         ?
-N                            ?
-p '-a 1 '                    ? -a is not a blastn option
-c                            ?
-v                            ?
-3  not defined in help       ?

[lieven.sterck]

nice error reporting !

OK, the problem with that specific sequence is the presence of that stretch of di-nucleotide repetition (TA) ... there blaster software can't handle this kind of very repetitive stretches. The best you can do is to let the REPET team know (@URGI) , it should not come as a surprise as they should be ware of this problem. As far as I know they try to fix it in the next release of REPET (blaster)

Also don't worry to much about the inconsistencies in the parameters between their blaster and blastn . They are for some understandable reason still referring to an old blast version ....