Convert the single-cell CNA profile based on the genomic bins locations to the one based on genes (here gene IDs).

This is a downstream util for the Navin lab's in-house pipeline CNA_pipeline v1.3 which uses hg19 (GRCh37).

1> CNA file across single cells generated by the CNA_pipeline v1.3.

For example:

chrom   chrompos    abspos  C1_S1_L001
1   977836  977836  0.726334447999411
1   1200863 1200863 0.726334447999411
1   1455238 1455238 0.726334447999411
2   0   249250621   1.60793145836484
2   237131  249487752   1.60793145836484
2   454245  249704866   1.60793145836484

Chrom 23 = X and 24 = Y.

2> Genomic locations of Varbin generated by the CNA_pipeline v.1.3.

This is a strange coordination with a mixed scheme. For example:

bin.chrom   bin.start   bin.end bin.length  gene.count  cgi.count   dist.telomere   gc.content
chr1    0   977835  977836  23  30  919083  0.45760009212891
chr1    977836  1200862 223027  11  41  1146783 0.611899007743492
chr1    1200863 1455237 254375  18  33  1371964 0.592286977886868
chr1    1455238 1758056 302819  11  35  1699329 0.545698915854107

Chrom is named as chr1, chr2, ..., chr22, chrX, chrY.

3> Genomic locations for genes (gene_id) downloaded from Ensembl BioMart (Ensembl-75 for the latest hg19). These columns are selected: Chromosome Name, Gene Start (bp), Gene End (bp), Ensembl, Gene ID, Associated Gene Name, Strand.

For example:

Chromosome Name Gene Start (bp) Gene End (bp)   Ensembl Gene ID Associated Gene Name    Strand
1   50902700    50902978    ENSG00000271782 RP5-850O15.4    -1
1   103817769   103828355   ENSG00000232753 RP11-347K2.1    1
1   50927141    50936822    ENSG00000225767 RP5-850O15.3    1

Snakemake, bedops, Python-3

0. Examine the inputs format.
   1. Does CNA file have at least three columns with header: chrom chrompos abspos C1_S1_L001, for example. The possible names chromosome 23 and 24 would be changed to X and Y internally by this Snakemake.
   2. Does the varbin reference have at least 3 columns with header: bin.chrom bin.start bin.end. And is the coordination in a mixed style: [0, 0]?
   3. Does the Ensembl follows the format as described before?
1. Prepare a run.config file to save all the required inputs. The template of the config is available at the example folder.
2. Run this command snakemake -s Snakefile.snakefile --configfile run.config all at terminal.

Only human genome is supported because internally the Snakemake file explicitly changes the chromosome 23 and 24 to X and Y, respectively.