Re: [Breakdancer-help] issue with BreakdancerMAX and MINI

[B.Panda <b....@st...>]

Ken,

Thank you for getting back quick. There is a large discrepancy. This is what we get:

1. 96,595 lines in the output file when we run whole genome (with -l (for illumina mate-pair long insert library) and -g option for bed file creation)
 2. 480 lines in the output file when we run with -t option on the same input file along with -l and -g options
3. ~37,000 lines in the output file when we run with the same input file for individual chromosomes with -o option and adding all the individual chromosome lines

Its clear that the whole genome does spit out something more than the -t option (trans-chromosomal changes) plus the individual chromosome ones (intra-chromosomal changes) but we don't know what that is.
 
Could you advice?

Binay

On Jul 17, 2010, at 5:43 AM, Ken Chen wrote:

> Hi Swetansu,
> 
>    It is possible that these don't quite add up, but I expect a large degree of consistency and that the discrepant calls have low scores or few number of supporting reads (e.g., 2, 3).  Could you let me know if that is the case?  Thanks.
> 
> -Ken
> 
> On 7/17/10 7:23 AM, Swetansu Pattnaik wrote:
>> Hey Ken,
>> The bam2cfg.pl was run successfully to create the cfg file.
>> However,  we are encountering certain discrepancies in the number of
>> predicted Structural Variations when we run  the whole genome versus
>> when we run it chromosomewise using the -o option of
>> BreakDancerMax.pl. More specifically, the -o option which yields
>> intrachromosomal variations and the -t option which works exclusively
>> of -o to predict the whole genome translocations does not add up to
>> the number of variations predicted by the program for whole genome
>> (perl BreakdancerMax.pl -l configfile.cfg).
>> Please let me know if you need extra info.
>> Thanks
>> Swetansu
>> PS: Our data is Illumina long insert (mate-pair) library.
>> 
>> 
>> On Sat, Jun 12, 2010 at 10:29 AM, Swetansu Pattnaik<swe...@gm...>  wrote:
>>   
>>> Thanks for your suggestions. I ll give you an update soon.
>>> Thanks
>>> Swetansu
>>> 
>>> On Sat, Jun 12, 2010 at 2:02 AM, Ken Chen<kc...@wa...>  wrote:
>>>     
>>>> Hi Swtansu,
>>>> 
>>>> 1. You can try to download a sample data set from here.
>>>> ftp://genome.wustl.edu/private/285445973646655/BreakDancer_sample_data/
>>>> and test BreakDancerMax.pl on it.
>>>> 2. If bam2cfg.pl fails, you can try to manually create a configuration file.
>>>>  Please read the README file for more details.
>>>> 3. BreakDancerMini.pl will become obsolete.  Instead, I would recommend you
>>>> to use pindel.  It will accurately cover small indels.
>>>> 4. Please consider subscribe to bre...@li... if
>>>> interested, we will be providing update there.
>>>> 
>>>> Thanks.
>>>> 
>>>> -Ken
>>>> 
>>>> 
>>>> On 6/11/10 3:03 PM, Swetansu Pattnaik wrote:
>>>>       
>>>>> Hi Ken,
>>>>> I have been trying to work on some Illumina reads using Breakdancer.
>>>>> As the manual suggests, the input for your program is a BAM file,
>>>>> which could either be generated from MAQ(map file in BAM format) or
>>>>> SAMtools(Novoalign SAM output converted into BAM). However, I have not
>>>>> been able to get ahead with this.
>>>>> Also the bam2cfg.pl script has issues.
>>>>> As of now I would need help with the following:
>>>>> 1>    Please send me a sample input file to work with.
>>>>> 2>    Format specifications of input file. Is there an alternative to
>>>>> bam2cfg.pl script?
>>>>> Thanks
>>>>> Swetansu
>>>>> 
>>>>>         
>>>> 
>>>>       
>>>     
>