diff --git a/.gitignore b/.gitignore
index 5c70e7d2ce9bc2a70a5602b76fc50fead3c79b81..bc34c3fdac7b6712b5da848564d95a6bcee0483d 100644
--- a/.gitignore
+++ b/.gitignore
@@ -14,4 +14,5 @@ KAPAC/doIt.sh
KAPAC/RESULTS/
KAPAC/RESULTS.log
KAPAC/DATA/kmer_counts_all.tsv
-
+# exclude the gitignore as well
+.gitignore
diff --git a/KAPAC/KAPAC.R b/KAPAC/KAPAC.R
index ad4cb88ea736551a71a856f9439b18ede0aea4d7..5e7e518c747962842fbfa3344552b946b484ec64 100755
--- a/KAPAC/KAPAC.R
+++ b/KAPAC/KAPAC.R
@@ -30,7 +30,7 @@ option_list <- list(
make_option(c("--expression_pseudocount"), default=1.0, action="store", type="double", help="The pseudocount that should be add to the expression values prior going to log-space."),
make_option(c("--consider_excess_counts_only"), default=TRUE, action="store", type="logical", help="Background correction: should the motif counts be corrected by the number of motifs that are expected to be found per chance."),
make_option(c("--considered_region_length"), action="store", type="double", help="Background correction: The length of the regions in which the k-mers are counted (needed for background correction)."),
- make_option(c("--min_kmer_abundance_fraction"), action="store", type="double", help="The fraction of (all) poly(A) sites that needs to contain a specific k-mer in order to consider the k-mer. E.g. 0.01 would require that a k-mer is found in at least 1% of all sites."),
+ make_option(c("--min_kmer_abundance_fraction"), default=0.01, action="store", type="double", help="The fraction of (all) poly(A) sites that needs to contain a specific k-mer in order to consider the k-mer. E.g. 0.01 would require that a k-mer is found in at least 1% of all sites."),
make_option(c("--number_of_randomized_runs"), default=1000, action="store", type="double", help="The number of runs done with randomized expression to k-mer count associations."),
make_option(c("--pas_overlap_col"), action="store", type="character", help="The column name of the column in the 'expression_matrix' that contains 'OK' for non overlapping poly(A) sites and 'OVERLAP' otherwise."),
make_option(c("--verbose"), default=TRUE, action="store", type="logical", help="Should the script be verbose (reporting detailed infos on what is done)."))
@@ -78,7 +78,7 @@ docs_80_dashes = "--------------------------------------------------------------
# _____________________________________________________________________________
# -----------------------------------------------------------------------------
-# Create nucleotides frequency matrix
+# Create the nucleotides frequency vector
# -----------------------------------------------------------------------------
# In case we want to background correct, we will use the following
# frequencies and report it to the user
@@ -1584,63 +1584,65 @@ write_incl_rownames(data=results.single_kmer_per_run.sorted[,cols_to_write],
# ---------------------------------------------------------------------------
# Since in the sitecount matrices we only consider k-mers that have been found
# in the defined region, it can be that we do not have every k-mer represented
-# in the sitecount matrices. Such k-mers we have to add by hand setting them
-# to 0.0 counts
+# in the sitecount matrix. This can also happen due to filtering out k-mers
+# based on their abundance (--min_kmer_abundance_fraction). For such k-mers
+# we still might want to have output files. One can create them as shown
+# below.
# ---------------------------------------------------------------------------
-if (opt$create_files_for_each_motif & (opt$selected_motifs != "all"))
-{
- # get k-mers for which a result is requested, but which was not considered
- # in the fit (maybe also because there do not exist counts in the currently
- # investigated region).
- missing_kmers = setdiff(selected_kmers, colnames(kapac.result[["Ns"]]))
-
- if (length(missing_kmers) > 0)
- {
- if (opt$verbose) {
- message(paste(docs_80_unerlines, sep=""))
- message(paste(docs_80_dashes, sep=""))
- message(paste("[INFO] WRITING FILES FOR K-MERS THAT ARE REQUESTED (--selected_kmers) BUT DO NOT HAVE COUNTS.", sep=""))
- }
-
- # -----------------------------------------------------------------------------
- # create a matrix that contains only 0.0 as activities and also as deltas
- # HINT: if a k-mer is not found at all within a region, it should have an
- # activity of 0.0 and we are also 100% about this (delta=0.0).
- template_matrix = t(kapac.result$Ahat)
- template_matrix[,1] = 0.0
-
- ##kmer = missing_kmers[1]
- for (kmer in missing_kmers)
- {
- # create the k-mer matrix
- kmer_matrix = template_matrix
- colnames(kmer_matrix) = kmer
-
- # -----------------------------------------------------------------------------
- # create the results directory
- path_kmer_results_dir = paste(opt$results_dir, kmer, sep="/")
- dir.create(path_kmer_results_dir, showWarnings=FALSE, recursive=TRUE)
-
- # prepare a matrix containing the z-scores (sorted)
- overall_z_scores = as.matrix(c(0.0),nrow=1, ncol=1)
- rownames(overall_z_scores) = kmer
- colnames(overall_z_scores) = 'z_score'
-
- # -----------------------------------------------------------------------------
- # write out the zscores
- write_incl_rownames(data=overall_z_scores,
- col_name='pwm',
- filename=paste(path_kmer_results_dir, '/zscore', sep=''))
-
- # write out the activities
- write_incl_rownames(data=kmer_matrix,
- col_name='sample_id',
- filename=paste(path_kmer_results_dir, '/activities', sep=''))
-
- # write out the activities
- write_incl_rownames(data=kmer_matrix,
- col_name='sample_id',
- filename=paste(path_kmer_results_dir, '/deltas', sep=''))
- }
- }
-}
+# if (opt$create_files_for_each_motif & (opt$selected_motifs != "all"))
+# {
+# # get k-mers for which a result is requested, but which was not considered
+# # in the fit (maybe also because there do not exist counts in the currently
+# # investigated region).
+# missing_kmers = setdiff(selected_kmers, colnames(kapac.result[["Ns"]]))
+#
+# if (length(missing_kmers) > 0)
+# {
+# if (opt$verbose) {
+# message(paste(docs_80_unerlines, sep=""))
+# message(paste(docs_80_dashes, sep=""))
+# message(paste("[INFO] WRITING FILES FOR K-MERS THAT ARE REQUESTED (--selected_kmers) BUT DO NOT HAVE COUNTS.", sep=""))
+# }
+#
+# # -----------------------------------------------------------------------------
+# # create a matrix that contains only 0.0 as activities and also as deltas
+# # HINT: if a k-mer is not found at all within a region, it should have an
+# # activity of 0.0 and we are also 100% about this (delta=0.0).
+# template_matrix = t(kapac.result$Ahat)
+# template_matrix[,1] = 0.0
+#
+# ##kmer = missing_kmers[1]
+# for (kmer in missing_kmers)
+# {
+# # create the k-mer matrix
+# kmer_matrix = template_matrix
+# colnames(kmer_matrix) = kmer
+#
+# # -----------------------------------------------------------------------------
+# # create the results directory
+# path_kmer_results_dir = paste(opt$results_dir, kmer, sep="/")
+# dir.create(path_kmer_results_dir, showWarnings=FALSE, recursive=TRUE)
+#
+# # prepare a matrix containing the z-scores (sorted)
+# overall_z_scores = as.matrix(c(0.0),nrow=1, ncol=1)
+# rownames(overall_z_scores) = kmer
+# colnames(overall_z_scores) = 'z_score'
+#
+# # -----------------------------------------------------------------------------
+# # write out the zscores
+# write_incl_rownames(data=overall_z_scores,
+# col_name='pwm',
+# filename=paste(path_kmer_results_dir, '/zscore', sep=''))
+#
+# # write out the activities
+# write_incl_rownames(data=kmer_matrix,
+# col_name='sample_id',
+# filename=paste(path_kmer_results_dir, '/activities', sep=''))
+#
+# # write out the activities
+# write_incl_rownames(data=kmer_matrix,
+# col_name='sample_id',
+# filename=paste(path_kmer_results_dir, '/deltas', sep=''))
+# }
+# }
+# }