# 
# pipeline2: after running pipeline1, maps the ".1pct.xls" files matching each experiment type
#            - computes overlapping peaks
#            - finds gene counts
#

#######################################
###### the types of files used ########
#######################################
type annotfile;		# annotation file
type listfile;		# simple list, or at least constant-column numberings
type file;		# generic file

#######################################
##### list of apps and processes ######
#######################################

app (annotfile output) comppeaks( listfile list[], int max_dist ){
	comppeaks @filename(list) max_dist @output;
}

app (listfile output) genecount( annotfile input, int gene_type, int divide ){
	genecount @input gene_type divide @output;
}

(annotfile output) compare_peaks ( string expt_type ) {
	listfile list[]<filesys_mapper; location="/export3/chip-seq/analyses/mmc_analyses", prefix=expt_type, suffix="1pct.xls">;
	output = comppeaks( list, 200 );
}


#######################################
############ pipeline #################
#######################################

# define all arrays to use; in all, should generate 3*(number of quest maps)+(number of expt maps)
file expts[]<ext; exec="exptmapper.sh">;		# input: list of unique experiment prefixes
annotfile overlaps[]<ext; exec="exptnamesmapper.sh">;	# output/input: file names for overlaps
listfile genes[]<ext; exec="exptgenesmapper.sh">;	# output: gene counts for overlaps

# compute overlaps and gene counts for each experiment type
foreach f, i in expts {
	overlaps[i] = compare_peaks( @filename(f) );
}

foreach g, j in overlaps {
	genes[j] = genecount( g, 0, 1 );
}