Initial commit of some kmer utilities.

there are two utilties included.

one is kmer_frequency_per_sequence,
which outputs a (m x n) matrix where m is the sequence, and n is the
frequency of that nmer to occur in the given sequence.

the other tool is kmer_total_count, which counts kmers for the total
file, not just one sequence

1 files changed, 49 insertions, 0 deletions

diff --git a/kmer_total_count.c b/kmer_total_count.c
new file mode 100644
index 0000000..4d1ab87
--- /dev/null
+++ b/kmer_total_count.c
@@ -0,0 +1,49 @@
+// Copyright 2013 Calvin Morrison
+#include <stdio.h>
+#include <math.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <stdint.h>
+
+#include "kmer_utils.h"
+
+int main(int argc, char **argv) {
+
+  char *line = NULL;
+  size_t len = 0;
+  ssize_t read;
+  unsigned int i = 0;
+
+  if(argc != 3) {
+    printf("Please supply a filename, and only a filename\n");
+    exit(EXIT_FAILURE);
+  }
+
+  FILE *fh = fopen(argv[1], "r" );
+  if(fh == NULL) {
+    fprintf(stderr, "Couldn't open: %s\n", argv[1]);
+    exit(EXIT_FAILURE);
+  }
+  
+  int kmer = atoi(argv[2]);
+  int width = (int)pow(4, kmer);
+
+  unsigned long long *counts = malloc((width+ 1) * sizeof(unsigned long long));
+  if(counts == NULL) 
+    exit(EXIT_FAILURE);
+
+  while ((read = getline(&line, &len, fh)) != -1) {
+    if(line[0] != '>')   {
+
+      for(i = 0; i < strlen(line) - kmer; i++) {
+        counts[convert_kmer_to_index(&line[i],kmer, width)]++;
+      }
+    }
+  }
+
+  for(i = 0; i < width; i++)
+    printf("%llu\n", counts[i]);
+
+
+  return EXIT_SUCCESS;
+}