C functions to detect IUPAC sequences

python/obitools3/dms/capi/obitypes.pxd

@@ -12,6 +12,8 @@ cdef extern from *:
 
 cdef extern from "encode.h" nogil:
     bint only_ATGC(const_char_p seq)
+    bint only_IUPAC_DNA(const_char_p seq)
+    bint is_a_DNA_seq(const_char_p seq)
 
 
 cdef extern from "obitypes.h" nogil:

src/encode.c

@@ -54,6 +54,66 @@ bool only_ATGC(const char* seq)
 }
 
 
+bool only_IUPAC_DNA(const char* seq)
+{
+	const char* c = seq;
+
+	while (*c)
+	{
+		if (!((*c == 'A') || \
+			  (*c == 'T') || \
+			  (*c == 'G') || \
+			  (*c == 'C') || \
+			  (*c == 'U') || \
+			  (*c == 'R') || \
+			  (*c == 'Y') || \
+			  (*c == 'S') || \
+			  (*c == 'W') || \
+			  (*c == 'K') || \
+			  (*c == 'M') || \
+			  (*c == 'B') || \
+			  (*c == 'D') || \
+			  (*c == 'H') || \
+			  (*c == 'V') || \
+			  (*c == 'N') || \
+			  (*c == 'a') || \
+			  (*c == 't') || \
+			  (*c == 'g') || \
+			  (*c == 'c') || \
+			  (*c == 'u') || \
+			  (*c == 'r') || \
+			  (*c == 'y') || \
+			  (*c == 's') || \
+			  (*c == 'w') || \
+			  (*c == 'k') || \
+			  (*c == 'm') || \
+			  (*c == 'b') || \
+			  (*c == 'd') || \
+			  (*c == 'h') || \
+			  (*c == 'v') || \
+			  (*c == 'n') || \
+			  (*c == '.') || \
+			  (*c == '-')))
+		{
+			return 0;
+		}
+		else
+		{
+			c++;
+		}
+	}
+	return 1;
+}
+
+
+bool is_a_DNA_seq(const char* seq)
+{
+	if (only_ATGC(seq))
+		return true;
+	return only_IUPAC_DNA(seq);
+}
+
+
 byte_t* encode_seq_on_2_bits(const char* seq, int32_t length)
 {
 	byte_t*  seq_b;

src/encode.h

@@ -78,6 +78,36 @@ enum
 bool only_ATGC(const char* seq);
 
 
+/**
+ * @brief Checks if there are only IUPAC DNA characters in a
+ *        character string.
+ *
+ * @param seq The sequence to check.
+ *
+ * @returns A boolean value indicating if there are only
+ *          IUPAC DNA characters in a character string.
+ *
+ * @since May 2017
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+bool only_IUPAC_DNA(const char* seq);
+
+
+/**
+ * @brief Checks if a character string can be read as a DNA sequence encoded
+ *        with ACGT or IUPAC characters (in capital letters or not).
+ *
+ * @param seq The sequence to check.
+ *
+ * @returns A boolean value indicating if the character string
+ *          can be read as a DNA sequence.
+ *
+ * @since May 2017
+ * @author Celine Mercier (celine.mercier@metabarcoding.org)
+ */
+bool is_a_DNA_seq(const char* seq);
+
+
 /**
  * @brief Encodes a DNA sequence with each nucleotide coded on 2 bits.
  *