Re: Converting a colour-space assembly to nucelotides

[Shaun Jackman]

Hi Ganga, (cc'ed to abyss-users)

> it gives the following errors
>
> ABySS 1.2.0

I'd recommend upgrading to ABySS 1.2.3 if possible.

To convert the colour-space sequences to nucleotides, add the cs=1
parameter to abyss-pe:
abyss-pe cs=1 ...

This conversion does not yet handle Ns in the colour-space contigs, and
ABySS's default is currently to scaffold using Ns. One option is to
disable scaffolding:
abyss-pe cs=1 OVERLAP_OPTIONS=--no-scaffold SIMPLEGRAPH_OPTIONS=--no-scaffold ...

Cheers,
Shaun

On Fri, 2010-09-10 at 05:50 -0700, Ganga Jeena wrote:
> Shaun,
> We are using new version of abyss to assembler SOLid data.
> and I need to know if theres any way ABySS converts color space data
> to bases .
>
> Also as mentioned on README of ABySS 1.2.3
> ABYSS_OPTIONS
> additional options to the single-end assembler, ABYSS
>
> cs convert colour-space contigs to nucleotide contigs
> following
> assembly
>
> but the ABYSS sinlge end is not accepting cs as any option moroever
> where should it be used if you could please give an example
>
>
>
> I did assembly of Solexa reads with abyss and it worked smoothly
> without any errors erupting .
> But its not that easy with SOLiD
>
> with a very small data of uptil 1000 reads it is running and giving
> result
> but for further reads
> it gives the following errors
>
> ABySS 1.2.0
> /opt/ABySS/bin/ABYSS -k21 f -o TTT
> Reading `f'
> Colour-space assembly
> ABYSS: AssemblyAlgorithms.cpp:81: void
> AssemblyAlgorithms::loadSequences(ISequenceCollection*, std::string):
> Assertion `isdigit(seq[0])' failed.
> Aborted

[Shaun Jackman]

Hi Ganga, (cc'ed to abyss-users)

The conversion of a colour-space assembly to nucleotides is the very
last step. All the intermediate files will be in colour space; the final
${name}-contigs.fa will be in nucleotides.

Please report the first 110 lines of the file named `F', and I'll help
you troubleshoot.

Cheers,
Shaun

On Fri, 2010-09-10 at 22:43 -0700, Ganga Jeena wrote:
> Dear Shaun,
>
> The conversion from color space contigs to bases is not taking place
> as evident form the following result files' content
>
> Command:
> abyss-pe cs=1 OVERLAP_OPTIONS=--no-scaffold
> SIMPLEGRAPH_OPTIONS=--no-scaffold k=11 n=5 in='A B' name=ME
>
>
> Files generated :
>
> ==> ME-1.adj <==
> 0 11 3 ; 0- ; 7+
> 1 12 12 ; 1- 5- 5+ ; 7+
> 2 11 25 ; 2- 2+ ; 2+ 2-
> 3 12 4 ; 3- ; 6+
> 4 11 56 ; 4- 4+ ; 4+ 4-
> 5 11 20 ; 1- 5- 5+ ; 1+ 5+ 5-
> 6 13 19 ; 3+ 7+ ;
> 7 12 20 ; 0+ 1+ ; 6+
>
> ==> ME-1.fa <==
> >0 11 3
> 23222222223
> >1 12 12
> 232222222222
> >2 11 25
> 11111111111
> >3 12 4
> 222322222232
> >4 11 56
> 00000000000
>
> ==> ME-1.indel <==
>
> ==> ME-3.adj <==
> 0 11 3 ; 0- ; 7+
> 1 12 12 ; 1- 5- 5+ ; 7+
> 2 11 25 ; 2- 2+ ; 2+ 2-
> 3 12 4 ; 3- ; 6+
> 4 11 56 ; 4- 4+ ; 4+ 4-
> 5 11 20 ; 1- 5- 5+ ; 1+ 5+ 5-
> 6 13 19 ; 3+ 7+ ;
> 7 12 20 ; 0+ 1+ ; 6+
>
> ==> ME-3.fa <==
> >0 11 3
> 23222222223
> >1 12 12
> 232222222222
> >2 11 25
> 11111111111
> >3 12 4
> 222322222232
> >4 11 56
> 00000000000
>
> ==> ME-3.hist <==
>
> ==> ME-bubbles.fa <==
>
>
>
>
> The ABYSS single run and abyss-pe are not working for reads more than
> 100
>
> and gives following error , here files F R have 200 reads.
>
>
>
>
>
>
> ABYSS -k11 -q3 --coverage-hist=coverage.hist -s ME-bubbles.fa -o
> ME-1.fa F R
> ABySS 1.2.0
> ABYSS -k11 -q3 --coverage-hist=coverage.hist -s ME-bubbles.fa -o
> ME-1.fa F R
> Reading `F'

> Colour-space assembly
> ABYSS: AssemblyAlgorithms.cpp:81: void
> AssemblyAlgorithms::loadSequences(ISequenceCollection*, std::string):
> Assertion `isdigit(seq[0])' failed.

> make: *** [ME-1.fa] Aborted
>
>
>
>
>
> Similar probelms erupts with read number above 100
>
>
> is there any way we could decipher why the script isnt working as
> desired for a larger dataset ?
>
> --Regards,
> Ganga Jeena

[Ganga Jeena]

The 'cs' option for conversion of color -space contigs to to bases is
working with abyss-pe, but not with single end assembly using ABYSS .
Is there any way to convert the result file of single end assembly in
color space to bases ?
Regards,
Ganga Jeena

On Sep 13, 10:28 pm, Shaun Jackman <sjack...@bcgsc.ca> wrote:
> Hi Ganga, (cc'ed to abyss-users)
>

> The conversion of a colour-spaceassemblyto nucleotides is the very

> > The ABYSSsinglerun and abyss-pe are not working for reads more than

> > 100
>
> > and gives following error , here files F R have 200 reads.
>
> > ABYSS -k11 -q3  --coverage-hist=coverage.hist -s ME-bubbles.fa  -o
> > ME-1.fa F R
> > ABySS 1.2.0
> > ABYSS -k11 -q3 --coverage-hist=coverage.hist -s ME-bubbles.fa -o
> > ME-1.fa F R
> > Reading `F'
> > Colour-spaceassembly
> > ABYSS: AssemblyAlgorithms.cpp:81: void
> > AssemblyAlgorithms::loadSequences(ISequenceCollection*, std::string):
> > Assertion `isdigit(seq[0])' failed.
> > make: *** [ME-1.fa] Aborted
>
> > Similar probelms erupts with read number above 100
>
> > is there any way we could decipher why the script isnt working as
> > desired for a larger dataset ?
>
> > --Regards,
> > Ganga Jeena
>

> > On Sat, Sep 11, 2010 at 2:29 AM, Shaun Jackman <sjack...@bcgsc.ca>

> >         > but the ABYSS sinlgeendis not accepting cs as any option

> >         moroever
> >         > where should it be used if you could please give an example
>

> >         > I didassemblyof Solexa reads with abyss and it worked