Re: [MUMmer-help] delta-filter parameters for out of order matches

[Adam P. <aph...@gm...>]

I think the problem is the -u option. Just don't use it. Since you are
mapping multiple BACs to the same chromosome, it will be throwing away
alignments from different BACs that map to the same position on the
chromosome. You don't want this. -g should automatically throw away
contained alignments (from the same BAC), so it should effectively filter
out the alignments you don't want.

Best,
-Adam


On Fri, Feb 13, 2015 at 12:15 AM, Surya Saha <ss...@co...> wrote:

> Hi Adam,
>
> I discovered something interesting doing this round of tests. Consider
> this excerpt from the .delta file produced by nucmer where BAC2:501-1001 is
> aligned out of order on CHR01.
> >CHR01 BAC2 20000 1500
> 183 683 501 1001 0 0 0
> 0
> 3001 3500 1 500 0 0 0
> 0
> 4001 4500 1001 1500 0 0 0
> 0
> >CHR01 BAC5 20000 5050
> 683 3500 51 2868 0 0 0
> 0
> 4001 5000 2869 3868 0 0 0
> 0
>
> Using delta-filter -g removes the out of order alignment so thats good.
> >CHR01 BAC2 20000 1500
> 3001 3500 1 500 0 0 0
> 0
> 4001 4500 1001 1500 0 0 0
> 0
> >CHR01 BAC5 20000 5050
> 683 3500 51 2868 0 0 0
> 0
> 4001 5000 2869 3868 0 0 0
> 0
>
> Now if I add the -u flag to remove the non-unique alignments (delta-filter
> -g -u 99), I expect none of the alignments to show up but the offending
> alignment reappears
> >CHR01 BAC2 20000 1500
> 183 683 501 1001 0 0 0
> 0
>
> The order of precedence is "-i -l -u -q -r -g -m -1" so -g should chuck
> this alignment out even if -u allows it. Am I missing something? Thanks
> much.
>
> -Surya
>
>
>
>
> On Thu, Feb 12, 2015 at 3:06 PM, Adam Phillippy <aph...@gm...>
> wrote:
>
>> Hi Surya,
>> Can you give me an example of an alignment (just the coords) that you
>> want to be filtered, but is left by delta-filter -g?
>>
>> -Adam
>>
>>
>> On Thu, Feb 12, 2015 at 12:24 AM, Surya Saha <ss...@co...> wrote:
>>
>>> Hi Adam,
>>>
>>> The delta-filter -g option with -u does not filter out the offending
>>> alignments in all cases. I think I will parse the delta file from nucmer to
>>> identify these out of order alignments unless there is a better option.
>>> Thank you for the prompt response!
>>>
>>> Best,
>>> Surya
>>> .
>>>
>>> On Wed, Feb 11, 2015 at 4:42 PM, Adam Phillippy <aph...@gm...>
>>> wrote:
>>>
>>>> Hi Surya,
>>>> See if the delta-filter -g option does what you want. That option will
>>>> enforce that the relative ordering of alignments on both the reference and
>>>> query is consistent.
>>>>
>>>> Best,
>>>> -Adam
>>>>
>>>>
>>>> On Wed, Feb 11, 2015 at 1:24 PM, Surya Saha <ss...@co...> wrote:
>>>>
>>>>> Hi Adam et al.,
>>>>>
>>>>> I am trying to use nucmer/delta-filter (v3.23) to align BACs to a
>>>>> chromosome with gaps (Ns) with a goal to find BACs that span gaps. I want
>>>>> to filter out any imperfect alignments. This also means any BAC that aligns
>>>>> out of order on the chromosome.
>>>>> Here is an example with out of order alignments (100% identity)
>>>>> BAC:100-200 aligns to CHR:500-600
>>>>> BAC:201-300 aligns to CHR:1001-1100
>>>>> BAC:301-400 aligns to 701-800
>>>>>
>>>>> I am trying to optimize delta-filter parameters for screening these
>>>>> out. I can screen out all alignments where a region of the chromosome is
>>>>> covered by multiple BACs with -u (99-100). Any ideas how to screen out the
>>>>> out of order alignments with delta-filter? Thanks!
>>>>>
>>>>>
>>>>> -Surya
>>>>>
>>>>>
>>>>> --
>>>>>
>>>>> Surya Saha
>>>>> Sol Genomics Network,
>>>>> Boyce Thompson Institute, Ithaca, NY, USA.
>>>>> Plant Pathology and Plant-Microbe Biology Section,
>>>>> Cornell University, NY, USA.
>>>>> http://www.linkedin.com/in/suryasaha
>>>>> https://twitter.com/SahaSurya
>>>>>
>>>>>
>>>>> ------------------------------------------------------------------------------
>>>>> Dive into the World of Parallel Programming. The Go Parallel Website,
>>>>> sponsored by Intel and developed in partnership with Slashdot Media,
>>>>> is your
>>>>> hub for all things parallel software development, from weekly thought
>>>>> leadership blogs to news, videos, case studies, tutorials and more.
>>>>> Take a
>>>>> look and join the conversation now. http://goparallel.sourceforge.net/
>>>>> _______________________________________________
>>>>> MUMmer-help mailing list
>>>>> MUM...@li...
>>>>> https://lists.sourceforge.net/lists/listinfo/mummer-help
>>>>>
>>>>>
>>>>
>>>
>>>
>>> --
>>>
>>> Surya Saha
>>> Sol Genomics Network,
>>> Boyce Thompson Institute, Ithaca, NY, USA.
>>> Plant Pathology and Plant-Microbe Biology Section,
>>> Cornell University, NY, USA.
>>> http://www.linkedin.com/in/suryasaha
>>> https://twitter.com/SahaSurya
>>>
>>
>>
>
>
> --
>
> Surya Saha
> Sol Genomics Network,
> Boyce Thompson Institute, Ithaca, NY, USA.
> Plant Pathology and Plant-Microbe Biology Section,
> Cornell University, NY, USA.
> http://www.linkedin.com/in/suryasaha
> https://twitter.com/SahaSurya
>