[Crac-bugs] CIGAR problem in CRAC

Linlin Yan (颜林林) yanll at mail.cbi.pku.edu.cn
Lun 13 Mai 09:56:53 CEST 2013


Hi Mike,

Thank you for the information. I will try it.

On Mon, May 13, 2013 at 3:00 PM, Mikaël Salson <mikael.salson at lifl.fr> wrote:
> Dear Linlin,
>
> A new version of CRAC is available. You can download it from here:
> https://gforge.inria.fr/frs/?group_id=3008&release_id=8050
>
> It addresses your issue and also implements the proposition you made (having
> gzipped output, you should use --gz option).
>
> Sorry for the delay and thanks a lot for your input!
> Mikael
>
>
> On 24/04/2013 00:07, Linlin Yan (颜林林) wrote:
>>
>> Hi Mikael,
>>
>> Thank you very much! It is very kind of you.
>>
>> The option "--all" will output multiple files, among which the .normal
>> and .almostNormal files are usually big enough for compressing to save
>> disk space.
>>
>> I am trying CRAC on some RNAseq data to find some de novo splicing
>> site and gene fusion events.
>>
>> Linlin
>>
>> On Tue, Apr 23, 2013 at 4:02 PM, Mikaël Salson <mikael.salson at lifl.fr>
>> wrote:
>>>
>>> Dear Linlin,
>>>
>>> Thanks a lot for your report.
>>> We identified the origin of the problem, and we will keep you informed
>>> as soon as we have a solution.
>>>
>>>
>>>> BTW, could you consider to export results in compressed format (.bam
>>>>   and .gz)?
>>>
>>>
>>> Thanks for the idea. We will see if we can implement that. I guess .gz
>>> output would be easier to implement (but maybe less convenient than bam).
>>>
>>>
>>>> Thank you for the excellent work! I really like the tool CRAC!
>>>
>>> Thank you! Just out of curiosity, for what purpose are you using CRAC?
>>>
>>> Best regards,
>>> Mikael
>>>
>>>
>>> On 21/04/2013 06:08, Linlin Yan (颜林林) wrote:
>>>>
>>>>
>>>> Hi,
>>>>
>>>> Even when I tried the latest CRAC (v1.2.4), the problem still exists
>>>> when I was trying to convert output.sam into BAM format. It
>>>> reported:
>>>>
>>>> [samopen] SAM header is present: 84 sequences. Line 10376251,
>>>> sequence length 98 vs 268435554 from CIGAR Parse error at line
>>>> 10376251: CIGAR and sequence length are inconsistent
>>>>
>>>> My command line is: crac -i /rd/user/yanll/data/crac/human_g1k_v37 -k
>>>> 22 -m 98 --nb-threads=4 -r ./00_data/H9_RNAseq_1.fq --sam
>>>> ./40_map_by_crac/H9.crac_v1_2_4.sam --all
>>>> ./40_map_by_crac/H9.crac_v1_2_4 ./00_data/H9_RNAseq_2.fq
>>>>
>>>> The bad result SAM line is:
>>>> HWI-ST483:306:C143CACXX:3:1106:18488:94443      129     12 20727519
>>>> 254     51S4294967295M26I22M    X       8789140 0
>>>>
>>>>
>>>> ACCATCACTATCATCACAATCACCACCACCATCATCGGCACCGTCATTATCATTATCATCACCATCACCACCACAACATTATCATCACCATCACCATC
>>>
>>>
>>>
>>> FFFHHHHHGGHHHHIJJJIJJJJJIHHGIIIJJJJHIJJJJHIIIIIFIJJGIIJJIHJIFHFEHFFFFCDDDDDDDCACCEDCDDCDCCDDCDD9
>>>>
>>>>
>>>>
>>>>
>>> XU:i:1  XD:i:1  XM:i:0  XN:i:0  XO:Z:11|-1,69871609     XQ:i:12
>>>>
>>>>
>>>> XC:i:2       XE:Z:0:0:BioUndetermined:50:[ambiguous case : no start
>>>> break]   XE:Z:1:1:Error:Ins:75:0.0744867:26:<snip>
>>>>
>>>> BTW, could you consider to export results in compressed format (.bam
>>>>   and .gz)?
>>>>
>>>> Thank you for the excellent work! I really like the tool CRAC!
>>>>
>>>> Best wishes!
>>>>
>>>> Linlin
>>>>
>>>> _______________________________________________ Crac-bugs mailing
>>>> list Crac-bugs at lists.gforge.inria.fr
>>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/crac-bugs
>>>>
>>>
>>
>



More information about the Crac-bugs mailing list