Skip to content

Prevent duplicate matches #5

Description

@hepcat72

There's a problem that has been an issue for a long time, and that is duplicates in the added feature columns. The same feature can get added as annotation to a data row multiple times. Here's an example:

Output:

>cat JS11_bamCoverage.bedgraph.nocov.simple
#featureProximity.pl Version 3.8
# Created: 11/2/2011
# Last modified: Wed Mar 15 16:13:59 2017
#Wed Mar 15 16:14:46 2017
#/usr/bin/perl /Users/rleach/pub/seqtools/featureProximity.pl -i deletions_unique.txt -f JS11_bamCoverage.bedgraph.nocov -r 0 -c 1 -b 2 -e 3 -a 1 -j 2 -k 3 -w 1
#datcol(1)	datcol(2)	datcol(3)	Distance	ftcol(1)
1	33574	33575		
1	36126	36141		
1	580578	580581		
1	2078797	2078798	0	1,1
1	2622551	2622908		
1	2652306	2652316	0	1
1	2653569	2653597		
1	2653683	2653691	0	1,1
1	2743286	2743305	0	1,1
1	2744893	2744896		
1	2745174	2745199	0	1
1	2850668	2850689		
1	2852896	2852903	0	1
1	3062741	3062792	0	1
1	3071674	3071683	0	1
1	3127342	3127343		
1	3178221	3178235	0	1,1
1	3217467	3217492	0	1,1
1	3223021	3223022		
2	26955	26956		
2	136633	136669	0	2
2	1021481	1021482		
2	1021487	1021488	0	2,2
2	1022198	1022199		
2	1334133	1334134	0	2,2
2	1776702	1777433		

Note the "1,1" and "2,2" feature annotations. Those are duplicate matches separated by commas. There is really only 1 such feature that overlaps the matching positions. Note the inputs below:

Inputs:

>cat JS11_bamCoverage.bedgraph.nocov
1	37318	37335	0
1	1895314	1895318	0
1	2078797	2078798	0
1	2410087	2410091	0
1	2652314	2652315	0
1	2653071	2653169	0
1	2653683	2653690	0
1	2743286	2743301	0
1	2745176	2745197	0
1	2851251	2851256	0
1	2852897	2852901	0
1	3060846	3060899	0
1	3061005	3061065	0
1	3061165	3061179	0
1	3062772	3062781	0
1	3066730	3066852	0
1	3071681	3071682	0
1	3072186	3072347	0
1	3110774	3110988	0
1	3112783	3112785	0
1	3117996	3118031	0
1	3121421	3121471	0
1	3178226	3178235	0
1	3178327	3178337	0
1	3217490	3217492	0
2	134711	134715	0
2	135204	135328	0
2	136639	136661	0
2	1021487	1021488	0
2	1334133	1334134	0
>cat deletions_unique.txt
1	33574	33575
1	36126	36141
1	580578	580581
1	2078797	2078798
1	2622551	2622908
1	2652306	2652316
1	2653569	2653597
1	2653683	2653691
1	2743286	2743305
1	2744893	2744896
1	2745174	2745199
1	2850668	2850689
1	2852896	2852903
1	3062741	3062792
1	3071674	3071683
1	3127342	3127343
1	3178221	3178235
1	3217467	3217492
1	3223021	3223022
2	26955	26956
2	136633	136669
2	1021481	1021482
2	1021487	1021488
2	1022198	1022199
2	1334133	1334134
2	1776702	1777433

I suspect that this issue has to do with separate loops that search for features with different orientations or searches different ways relative to the data coordinates. I recall having looked into this in the past and I remember that it was a difficult bug to resolve.

Metadata

Metadata

Assignees

No one assigned

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions