Oases: De novo transcriptome assembly of very short reads

[Symphysodon]

Thanks for your suggestions re. Oases: segmentation fault.

I have recompiled Oases, making sure to give the correct path to Velvet and specifying the same 'CATEGORIES=30' 'MAXKMERLENGTH=99' as for Velvet.

Unfortunately, I still get the same error. So, this error is indeed very reproducible .

I'm using a hash value of 57, which gave me the best N50 and largest contig out of a range of 51-99mers - from VelvetOptimiser and subsequent optimisation runs.

[Thorondor]

woah! how many reads you have and how long are they? Never choose a higher hash value then your minimum read length.

so you also get sementation fault with lower kmers like 43 or higher onces like 76?

[Symphysodon]

I have total of a little over 22 million 99-nucleotide reads. This is from four lanes of Illumina data.

I have just assembled one lane of data (5,522,680) and Oases seems to be able to cope with this, at a hash value of 57. So I guess the quantity of input data is a factor for the segmentation fault. I'll raise this with Marcel or Daniel.

Thanks all.

[lfaino]

dear all,
I have a problem with OASES. I run this :
cd ~/processing
velveth 25march/Oidium_fasta_RNAseq/assembly 25 -fasta -shortPaired 25march/Oidium_fasta_RNAseq/Oidium.fa
velvetg 25march/Oidium_fasta_RNAseq/assembly -read_trkg yes
oases 25march/Oidium_fasta_RNAseq/assembly -alignments yes -unused_reads yes

the problem is that wen I add some options to OASES (-alignments yes -unused_reads yes) it does not work at all.

Any suggestions?

Luigi

[Thorondor]

you need to give the insert length as parameter (-ins_length) as stated in the readme.

Which versions of velvet and oases you use?

did you compile oases with the same maxkmerlength and categories as used in velvet?

[schmima]

try to use the options also during velvetg:

velvetg ... -alignments yes -unused_reads yes -read_trkg yes

[lfaino]

dear Thorondor,
In velveth the parameters are CATEGORIES = 2 MAXKMERLENGTH = 31 which are the standard from the Makefile. In Oases I can not trace them back but the Makefile has the some parameters in it. I assume they are the same than.
About the answer of Schmima, it is possible to have these files in velvetg but the results can be different between the oases and velvetg.
any other suggestions?

Luigi

[schmima]

yes - the output files normally differ. Just thought that oases may need the velvet output from these files and modifies them (?).

btw - what are the errors you get? or in what way is it not working?

[Thorondor]

just call ./oases without parameters and the output will be like:

oases - De novo transcriptome assembler for the Velvet package
Version 0.1.20

Copyright 2009,2010 Daniel Zerbino (dzerbino@soe.ucsc.edu)
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

Compilation settings:
CATEGORIES = 27
MAXKMERLENGTH = 61

plus parameters you can set. same works for velveth or velvetg.

So you have the newest versions installed of velvet and oases and you have set the insert length only in oases?
Don't set any other parameters in velvetg then -read_trkg yes!

what files are there in you folder after running velveth and velvetg?

[Jenzo]

Dear Oases users,

I'm trying to compile Oases on a Ubuntu 10.10 virtual machine on our server. Directly after installing the OS, i just installed G++ for compiling. Velvet was compiled with no errors, but Oases gives me the following output:

[...]
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/transcript.c -o obj/transcript.o -I/home/jpreuss/Velvet/src
src/transcript.c:564: error: static declaration of ‘getTotalCoverage’ follows non-static declaration
/home/jpreuss/Velvet/src/graph.h:99: note: previous declaration of ‘getTotalCoverage’ was here
make: *** [obj/transcript.o] Error 1

Has anyone an idea why this error shows up, or what it means? I got Oases already running on Ubuntu 10.10, perhaps I'm missing another lib?

Thanks for any suggestions,
Jenzo

[Thorondor]

hmmm, never got a error like that.

you have velvet 1.1.03 and oases 0.1.20 ?

[Jenzo]

Right, I downloaded them a few hours ago..

Here's the complete output:

jpreuss@velvet-vm:~/Oases$ make 'VELVET_DIR=/home/jpreuss/Velvet' 'MAXKMERLENGTH=35'
rm -f obj/*.o obj/dbg/*.o 
cd /home/jpreuss/Velvet && make obj
make[1]: Betrete Verzeichnis '/home/jpreuss/Velvet'
cd third-party/zlib-1.2.3; ./configure; make; rm minigzip.o; rm example.o
Checking for gcc...
Building static library libz.a version 1.2.3 with gcc.
Checking for unistd.h... Yes.
Checking whether to use vs[n]printf() or s[n]printf()... using vs[n]printf()
Checking for vsnprintf() in stdio.h... Yes.
Checking for return value of vsnprintf()... Yes.
Checking for errno.h... Yes.
Checking for mmap support... Yes.
make[2]: Betrete Verzeichnis '/home/jpreuss/Velvet/third-party/zlib-1.2.3'
gcc -O3 -DUSE_MMAP   -c -o example.o example.c
gcc -O3 -DUSE_MMAP   -c -o adler32.o adler32.c
gcc -O3 -DUSE_MMAP   -c -o compress.o compress.c
gcc -O3 -DUSE_MMAP   -c -o crc32.o crc32.c
gcc -O3 -DUSE_MMAP   -c -o gzio.o gzio.c
gcc -O3 -DUSE_MMAP   -c -o uncompr.o uncompr.c
gcc -O3 -DUSE_MMAP   -c -o deflate.o deflate.c
gcc -O3 -DUSE_MMAP   -c -o trees.o trees.c
gcc -O3 -DUSE_MMAP   -c -o zutil.o zutil.c
gcc -O3 -DUSE_MMAP   -c -o inflate.o inflate.c
gcc -O3 -DUSE_MMAP   -c -o infback.o infback.c
gcc -O3 -DUSE_MMAP   -c -o inftrees.o inftrees.c
gcc -O3 -DUSE_MMAP   -c -o inffast.o inffast.c
ar rc libz.a adler32.o compress.o crc32.o gzio.o uncompr.o deflate.o trees.o zutil.o inflate.o infback.o inftrees.o inffast.o 
gcc -O3 -DUSE_MMAP -o example example.o -L. libz.a
gcc -O3 -DUSE_MMAP   -c -o minigzip.o minigzip.c
gcc -O3 -DUSE_MMAP -o minigzip minigzip.o -L. libz.a
make[2]: Verlasse Verzeichnis '/home/jpreuss/Velvet/third-party/zlib-1.2.3'
rm obj/*.o obj/dbg/*.o 
rm: Entfernen von „obj/dbg/*.o“ nicht möglich: Datei oder Verzeichnis nicht gefunden
make[1]: [cleanobj] Fehler 1 (ignoriert)
mkdir -p obj
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/tightString.c -o obj/tightString.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/run.c -o obj/run.o 
src/run.c: In function ‘main’:
src/run.c:212: warning: ignoring return value of ‘system’, declared with attribute warn_unused_result
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/splay.c -o obj/splay.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/splayTable.c -o obj/splayTable.o 
src/splayTable.c: In function ‘computeClearHSPs’:
src/splayTable.c:631: warning: ignoring return value of ‘fgets’, declared with attribute warn_unused_result
src/splayTable.c: In function ‘inputSequenceArrayIntoSplayTableAndArchive’:
src/splayTable.c:931: warning: ignoring return value of ‘fgets’, declared with attribute warn_unused_result
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/graph.c -o obj/graph.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/run2.c -o obj/run2.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/fibHeap.c -o obj/fibHeap.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/fib.c -o obj/fib.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/concatenatedGraph.c -o obj/concatenatedGraph.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/passageMarker.c -o obj/passageMarker.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/graphStats.c -o obj/graphStats.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/correctedGraph.c -o obj/correctedGraph.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/dfib.c -o obj/dfib.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/dfibHeap.c -o obj/dfibHeap.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/recycleBin.c -o obj/recycleBin.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/readSet.c -o obj/readSet.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/shortReadPairs.c -o obj/shortReadPairs.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/locallyCorrectedGraph.c -o obj/locallyCorrectedGraph.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/graphReConstruction.c -o obj/graphReConstruction.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/roadMap.c -o obj/roadMap.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/preGraph.c -o obj/preGraph.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/preGraphConstruction.c -o obj/preGraphConstruction.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/concatenatedPreGraph.c -o obj/concatenatedPreGraph.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/readCoherentGraph.c -o obj/readCoherentGraph.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/utility.c -o obj/utility.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/kmer.c -o obj/kmer.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/scaffold.c -o obj/scaffold.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/kmerOccurenceTable.c -o obj/kmerOccurenceTable.o 
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/allocArray.c -o obj/allocArray.o 
make[1]: Verlasse Verzeichnis '/home/jpreuss/Velvet'
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/oases.c -o obj/oases.o -I/home/jpreuss/Velvet/src
gcc -Wall -O3 -D MAXKMERLENGTH=35 -D CATEGORIES=2 -c src/transcript.c -o obj/transcript.o -I/home/jpreuss/Velvet/src
src/transcript.c:564: error: static declaration of ‘getTotalCoverage’ follows non-static declaration
/home/jpreuss/Velvet/src/graph.h:99: note: previous declaration of ‘getTotalCoverage’ was here
make: *** [obj/transcript.o] Fehler 1

[Thorondor]

okey. can't help here. :-/

Anyway you should try to ask for help over the oases mailing list: http://listserver.ebi.ac.uk/mailman/...fo/oases-users

zerbino is also active there and normally answers fast.

[Jenzo]

Thanks a lot. I will try to get an answer there :-) SOLVED:

Hey!
I got an answer via the Oases mailing list. The problem is, according to Sylvain Foret, that 'getTotalCoverage' was made public in Velvet 1.1.03, but not changed in Oases. One can use Velvet 1.1.02 instead or comment out the definition of 'getTotalCoverage' in transcript.c.
It is suggested that Oases will be fixed.

[chrisbala]

Hi everyone,

I'm using OASES for the first time.

Quick question:
Should I expect anything to be printed to output while OASES is running? (aka immediately after I start the program)

It seems to be running but nothing has been printed and I would love to get some feel good progress report... (to make sure something is actually happening in there!)

thanks!

[chrisbala]

answer received: I got some output after a few minutes. Looks like things are going...

[Thorondor]

depending on your number of reads it takes "hours" till it finishes.

5hours
30GB RAM
with around 70.000.0000 reads with average length 90

you can speed things up with the new parallel mode but didn't tried that out yet.

[oliviera]

dear Jenzo, dear All,

I have the same problem than you Jenzo: Oases crashes with the same error that you describe. I guess it is better to use the last release of velvet. Have you managed to use the last release or did you go back to the 1.1.02?
Thanx
Olivier

[Jenzo]

Dear Olivier,
I'm using Velvet 1.1.03 with Oases 0.1.21. It works perfect now..
Perhaps you try to get the latest version of both, recompile and run it again, it should work then :-)
Greets,
Jenzo

[oliviera]

Thanx for your fast reply,
I download and install without problem velvet 1.1.03 with
make 'MAXKMERLENGTH=57' 'BIGASSEMBLY=1' 'OPENMP=1'

By ther way I have 76bp paired end reads so I thought it is interesting to use MAXKMERLENGTH=57. Is it right ?

I then tried to install the last release oases 0.1.21 with
make 'VELVET_DIR=/data/shared/src/velvet/velvet_1.1.03/' 'MAXKMERLENGTH=57' 'BIGASSEMBLY=1' 'OPENMP=1'
This is not working.
When I remove the option 'OPENMP=1' (I thought to use it to turn on the multithread processing) then installation goes until the end.
So I guess their is something wrong with Oases and the multithread support. Or I do something wrong