RepeatModeler FastaDB::getSubstr error

[Bitar]

I have the same exact issue during round 5

Processing RECON family: 21802
- Saving elements to a file...
FastaDB::getSubstr - Error index out of bounds! (SeqID=gi|4517, offset=2237, length=327 actualSeqLen=2547)
at /programs/RepeatModeler/RepeatModeler line 1974.

[jomaco]

Hi,

Did you have any luck with this problem?

I have been experiencing it for some time, but the only way round it seems to be running RepeatModeler many many times until it finally (and seemingly randomly) works.

I should say the error I most recently saw (and I've seen this on a few occasions) was:

FastaDB::getSubstr - Error index out of bounds! (SeqID=gi|104228, offset=1865, length=60 actualSeqLen=1920) at /usr/users/celldev/cockerj/repeatmodeler2/RepeatModeler/RepeatModeler line 1624

I've also seen:

Program duration is 1271.0 sec = 21.2 min = 0.4 hr eleredef failed. Exit code 11

Now I know eleredef is something to do with RECON but really I don't what these errors mean. With the same input files and options sometimes RepeatModeler works and sometimes it doesn't.

How can I make sure RepeatModeler works every time without erroring often?

Thanks,

Jom.

[jomaco]

Hi,

Did you have any luck with this problem?

I have been experiencing it for some time, but the only way round it seems to be running RepeatModeler many many times until it finally (and seemingly randomly) works.

How can I make sure RepeatModeler works every time without erroring often?

Thanks,

Jom.

[klduong]

My RepeatModeler run ended abruptly with this entry in the run.out file

Processing RECON family: 1234
- Saving elements to a file...
FastaDB::getSubstr - Error index out of bounds! (SeqID=gi|95483, offset=1, length=142 actualSeqLen=141)
at /usr/local/RepeatModeler/RepeatModeler line 1523

RepeatModeler was on round 5, and was doing just fine processing RECON families until it got this this set. Has anyone found a solution to this error?

I'm using RepeatModeler 1.0.5 with RMBlast. The only changes I have made to the program are as described here http://seqanswers.com/forums/showthread.php?t=24159, but those did not affect line 1523 or anything in the my $sequence = $batchDB->getSubstr( $seqName, $startOffset, $length ); command at that line, I think.

The RepeatMasker website has a patch for a similar error (FastaDB::substr - Error index out of bounds! at RepeatMasker line 4969), but it is for RepeatMasker's CrossmatchSearchEngine.pm module for Crossmatch. Can this patch also fix my RepeatModeler problem?