Add more upstream segments

Test/test.sh

@@ -24,33 +24,38 @@ printf "\n1-base overlaps:\n\n"
     test-overlaps.tsv
 ../filter-overlaps test-overlaps.tsv test-filtered.tsv \
     five_prime_utr three_prime_utr intron exon \
-    upstream1000 upstream10000 upstream100000 upstream-beyond
+    upstream1000 upstream10000 upstream100000 upstream200000 upstream300000 \
+    upstream400000 upstream500000 upstream600000 upstream700000 upstream800000 upstream-beyond
 
 printf "\n20%% peak overlaps:\n\n"
 ../peak-classifier --min-peak-overlap 0.2 test.bed.xz \
     $gff test-peak-20-overlaps.tsv
 ../filter-overlaps test-peak-20-overlaps.tsv test-peak-20-filtered.tsv \
     five_prime_utr three_prime_utr intron exon \
-    upstream1000 upstream10000 upstream100000 upstream-beyond
+    upstream1000 upstream10000 upstream100000 upstream200000 upstream300000 \
+    upstream400000 upstream500000 upstream600000 upstream700000 upstream800000 upstream-beyond
 
 printf "\n20%% GFF feature overlaps:\n\n"
 ../peak-classifier --min-gff-overlap 0.2 test.bed.xz \
     $gff test-gff-20-overlaps.tsv
 ../filter-overlaps test-gff-20-overlaps.tsv test-gff-20-filtered.tsv \
     five_prime_utr three_prime_utr intron exon \
-    upstream1000 upstream10000 upstream100000 upstream-beyond
+    upstream1000 upstream10000 upstream100000 upstream200000 upstream300000 \
+    upstream400000 upstream500000 upstream600000 upstream700000 upstream800000 upstream-beyond
 
 printf "\n20%% either peak or GFF feature overlaps:\n\n"
 ../peak-classifier --min-gff-overlap 0.2 --min-gff-overlap 0.2 \
     --min-either-overlap test.bed.xz \
     $gff test-either-20-overlaps.tsv
 ../filter-overlaps test-either-20-overlaps.tsv test-either-20-filtered.tsv \
     five_prime_utr three_prime_utr intron exon \
-    upstream1000 upstream10000 upstream100000 upstream-beyond
+    upstream1000 upstream10000 upstream100000 upstream200000 upstream300000 \
+    upstream400000 upstream500000 upstream600000 upstream700000 upstream800000 upstream-beyond
 
 printf "\nMidpoints only:\n\n"
 ../peak-classifier --midpoints test.bed.xz $gff \
     test-midpoint-overlaps.tsv
 ../filter-overlaps test-midpoint-overlaps.tsv test-filtered.tsv \
     five_prime_utr three_prime_utr intron exon \
-    upstream1000 upstream10000 upstream100000 upstream-beyond
+    upstream1000 upstream10000 upstream100000 upstream200000 upstream300000 \
+    upstream400000 upstream500000 upstream600000 upstream700000 upstream800000 upstream-beyond

filter-overlaps.c

@@ -136,8 +136,8 @@ int     filter_overlaps(const char *overlaps_file, const char *output_file,
 
     printf("Total unique peaks: %zu\n", unique_peaks);
     for (c = 0; features[c] != NULL; ++c)
-	printf("Overlaps with %-20s: %7zu (%2zu%%)\n", features[c],
-		feature_overlaps[c], 100 * feature_overlaps[c] / unique_peaks);
+	printf("Overlaps with %-20s: %7zu (%3.1f%%)\n", features[c],
+		feature_overlaps[c], 100.0 * feature_overlaps[c] / unique_peaks);
     return EX_OK;
 }
 

peak-classifier.c

@@ -36,7 +36,7 @@ int     main(int argc,char *argv[])
 	    *gff_stream,
 	    *intersect_pipe;
 	    // Default, override with --upstream-boundaries
-    char    *upstream_boundaries = "1000,10000,100000",
+    char    *upstream_boundaries = "1000,10000,100000,200000,300000,400000,500000,600000,700000,800000",
 	    *p,
 	    cmd[PEAK_CMD_MAX + 1],
 	    *redirect_overwrite,
@@ -46,7 +46,8 @@ int     main(int argc,char *argv[])
 	    *end,
 	    *gff_stem,
 	    augmented_filename[PATH_MAX + 1],
-	    sorted_filename[PATH_MAX + 1];
+	    sorted_filename[PATH_MAX + 1],
+	    *sort;
     bool    midpoints_only = false;
     bl_bed_t   bed_feature;
     struct stat     file_info;
@@ -150,9 +151,15 @@ int     main(int argc,char *argv[])
     else
     {
 	// LC_ALL=C makes sort assume 1 byte/char, which improves speed
+	// gsort is faster than other implementations, so use it if
+	// available
+	if ( system("which gsort") == 0 )
+	    sort = "gsort";
+	else
+	    sort = "sort";
 	snprintf(cmd, PEAK_CMD_MAX, "env LC_ALL=C grep -v '^#' %s | "
-		"sort -n -k 1 -k 2 -k 3 > %s\n",
-		augmented_filename, sorted_filename);
+		"%s -n -k 1 -k 2 -k 3 > %s\n",
+		augmented_filename, sort, sorted_filename);
 	fputs("Sorting...\n", stderr);
 	if ( (status = system(cmd)) != 0 )
 	{