Filtering pileup from site lists

Question

[Eliran Turgeman]

I want to write a script that filters a pileup file from a site lists file.
As an input I get a reference genome, pileup and site lists files.

Example of an output for this script:
Pileup File :

NC_000001.10    11456   A   0   *   *  
NC_000001.10    11457   C   0   *   *  
NC_000001.10    11458   T   0   *   *  
NC_000002.11    11460   G   1   ,   E  
NC_000002.11    13914   G   1   ,   E   
NC_000003.11    10503990    C   1   ,   E

Reference File (After using grep by '>'):

>NC_000001.10 Homo sapiens chromosome 1, GRCh37.p13 Primary Assembly
>NT_113878.1 Homo sapiens chromosome 1 unlocalized genomic contig, GRCh37.p13 Primary Assembly
>NT_167207.1 Homo sapiens chromosome 1 unlocalized genomic contig, GRCh37.p13 Primary Assembly
>NC_000002.11 Homo sapiens chromosome 2, GRCh37.p13 Primary Assembly
>NC_000003.11 Homo sapiens chromosome 3, GRCh37.p13 Primary Assembly

Lists File:

chr1    11456   C1orf186    -   intronic    intronic    no  no  N   N   N
chr1    11457   C1orf186    -   intronic    intronic    no  no  N   N   N
chr2    13914   intergenic  -   intergenic  intergenic  no  no  N   N   N
chr7    30504355    NOD1    -   intronic    intronic    no  no  N   N   N
chr3    10503990    SSX2IP  -   Syn Gln->Gln    no  no  N   N   N
chr1    13131320    MEF2A   +   intronic    intronic    no  no  N   N   N

Output File:

NC_000001.10    11456   A   0   *   *
NC_000001.10    11457   C   0   *   *
NC_000002.11    13914   G   1   ,   E
NC_000003.11    10503990    C   1   ,   E

So for every refseq in the pileup file I would convert it to the matching chromose<id> in which is found in the reference file, and then overlap it with my lists file so I would get the lines with the same chromose<id> and position.

The problem is that while there are many refseq coordinates that belong to chr1 for example, alignment to each of them will give a position, with respect to the specific coordinate. But on the list there is a full, single, chr1.

I need to be able to compare between:
* a pileup file that was created based on a refseq reference genome
* a file with editing sites that was downloaded from RADAR (we may give the file)
Maybe there is a more appropriate reference genome file or that there is a way to convert the refseq positions according to the list

[Pierre Lindenbaum]

convert your pileup to bed using awk and then use bedtools intersect

[Eliran Turgeman]

Could you please elaborate on how exactly I would need to use bedtools intersect?