[Bioclusters] mpiBLAST and NCBI Blast? BioseqFindFunc: couldn't uncache

[Dan Bolser]

On Wed, 8 Sep 2004, Joe Landman wrote:

>Dan Bolser wrote:
>
>>formatdb -i sequences.fa -o T
>>blastall -d sequences.fa -p blastp -i domain_sequences.fa  -m 8 -a4 > results
>>  
>>
>
>...
>
>
>>I use -m 8 because I don't care about the alignment, but perhaps that
>>would give clues to the error.
>>
>>If you like I can re-run the problem sequences with -m 7 and send along
>>some results.
>>  
>>
>
>Try omitting the -m # entirely, or use -m 7 .

Well... 

As a first test I tried re-running the problem sequences again. The only
difference being that now they are in the format produced by fastacmd
(which I used to pull them out of the original fasta database).

There were no errors during this run. When I format the original results
for these sequences to match the new results (with the new fasta format)
their was no 'diff' in the results...

This was over 92 sequences which gave this error one or more times during 
blastall.

i.e. the output of blastall -m 8 is not affected by this error.


Pulling out the problem sequences in their original format (without using
fastacmd) is somewhat tricky, so I only did this for one sequence.
Re-running this sequence in the same way as before gave no error,
suggesting that the error is not actually related to any particular
sequence or to the format of the fasta header. It appears to be sporadic
with no effect on the results.

Hmmm....


>
>
>>Cheers,
>>Dan.
>>  
>>
>
>
>