Re: [wgs-assembler-users] assertion fail in buildUnitigs step

[Walenz, B. <bw...@jc...>]

Hi-

I’ll look into the crash; it’s a strange one for sure.  Without being able to reproduce it here, I don’t predict much success.  To move the assembly forward, you can switch to the ‘bogart’ unitigger.   Remove the 4-unitigger and *tigStore directories before restarting.

I’d guess that your disk usage is from using 25% error.  The defaults of ovl=0.06 utg=0.03 work well for bacteria.

You can also try sampling down the reads to 50x-100x coverage with ‘fastqSample’.  46m reads * 100bp = 4.6g sequence = 1000x coverage?

b
--
Brian Walenz
Senior Software Engineer
J. Craig Venter Institute


On 4/6/12 12:17 PM, "Ishwor Thapa" <ish...@pk...> wrote:

Hi,

I am trying to use celera7 for an assembly of paired ends reads from a Bacterial Genome. It ran upto Unitigger step and it aborted due to assertion fail in buildUnitigs step.


Below is the content of 4-unitigger/unitigger.err

Bubble popping        = on
Intersection breaking = on
Bad mate threshold    = -7
Error threshold       = 0.250 (25.000%)
Error limit           = 4.500 errors

sizeof(ufPath)        = 24

FragmentInfo()-- Loading fragment information deleted:    55232 active:  9944768
FragmentInfo()-- Loading fragment information deleted:   113524 active: 19886476
FragmentInfo()-- Loading fragment information deleted:   188235 active: 29811765
FragmentInfo()-- Loading fragment information deleted:   363003 active: 39636997
FragmentInfo()-- Loaded 45918788 alive fragments, skipped 401422 dead fragments.
FragmentInfo()-- Saving fragment information to cache '/scratch/ithapa/celera/first/4-unitigger/UNK-2.fragmentInfo
'
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.001.bestoverlapgraph-containments.lo
g'
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.002.bestoverlapgraph-dovetails.log'
BestOverlapGraph()-- Saving overlap graph to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.bog'.
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.003.ChunkGraph.log'
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.004.buildUnitigs.log'
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.005.bubblePopping.log'
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.006.intersectionBreaking.log'
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.007.placeContains.log'
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.008.placeZombies.log'
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.009.bubblePopping.log'
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.010.libraryStats.log'
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.011.evaluateMates.log'
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.012.moveContains1.log'
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.013.splitDiscontinuous1.log'
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.014.splitBadMates.log'
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.015.splitDiscontinuous2.log'
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.016.moveContains2.log'
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.017.splitDiscontinuous3.log'
setLogFile()-- Now logging to '/scratch/ithapa/celera/first/4-unitigger/UNK-2.018.output.log'
buildUnitigs: AS_BOG_Outputs.cc:116: void UnitigGraph::writeIUMtoFile(char*, char*, uint32, bool): Assertion `utg-
>getLength() > 0' failed.
Aborted

I have the spec file as follows:
utgErrorRate=0.25
utgErrorLimit=4.5
cnsErrorRate=0.25
cgwErrorRate=0.25
ovlErrorRate=0.25
ovlHashBits=23
ovlHashBlockLength=300000000
ovlThreads = 2
ovlConcurrency = 2

Can you please let me know what might have gone wrong?

Another question I had is, if its normal to take more than 800GB disk space to reach to Unitigger step. Originally I had 12.4GB fastq file.


Thanks,
Ishwor