*/
package jalview.util;
-import jalview.datamodel.*;
+import jalview.datamodel.SequenceI;
+
+import java.util.ArrayList;
+import java.util.List;
/**
- * DOCUMENT ME!
- *
- * @author $author$
- * @version $Revision$
+ * Assorted methods for analysing or comparing sequences.
*/
public class Comparison
{
- /** DOCUMENT ME!! */
- public static final String GapChars = " .-";
+ private static final int EIGHTY_FIVE = 85;
+
+ private static final int TO_UPPER_CASE = 'a' - 'A';
+
+ public static final char GAP_SPACE = ' ';
+
+ public static final char GAP_DOT = '.';
+
+ public static final char GAP_DASH = '-';
+
+ public static final String GapChars = new String(
+ new char[]
+ { GAP_SPACE, GAP_DOT, GAP_DASH });
/**
* DOCUMENT ME!
* int
* @return float
*/
- public static float compare(SequenceI ii, SequenceI jj, int start, int end)
+ public static float compare(SequenceI ii, SequenceI jj, int start,
+ int end)
{
String si = ii.getSequenceAsString();
String sj = jj.getSequenceAsString();
int ilen = si.length() - 1;
int jlen = sj.length() - 1;
- while (jalview.util.Comparison.isGap(si.charAt(start + ilen)))
+ while (Comparison.isGap(si.charAt(start + ilen)))
{
ilen--;
}
- while (jalview.util.Comparison.isGap(sj.charAt(start + jlen)))
+ while (Comparison.isGap(sj.charAt(start + jlen)))
{
jlen--;
}
{
for (int j = 0; j < jlen; j++)
{
- if (si.substring(start + j, start + j + 1).equals(
- sj.substring(start + j, start + j + 1)))
+ if (si.substring(start + j, start + j + 1)
+ .equals(sj.substring(start + j, start + j + 1)))
{
match++;
}
{
for (int j = 0; j < jlen; j++)
{
- if (si.substring(start + j, start + j + 1).equals(
- sj.substring(start + j, start + j + 1)))
+ if (si.substring(start + j, start + j + 1)
+ .equals(sj.substring(start + j, start + j + 1)))
{
match++;
}
* @param s2
* SequenceI
* @return float
+ * @deprecated use PIDModel.computePID()
*/
+ @Deprecated
public final static float PID(String seq1, String seq2)
{
return PID(seq1, seq2, 0, seq1.length());
static final int caseShift = 'a' - 'A';
// Another pid with region specification
- public final static float PID(String seq1, String seq2, int start, int end)
+ /**
+ * @deprecated use PIDModel.computePID()
+ */
+ @Deprecated
+ public final static float PID(String seq1, String seq2, int start,
+ int end)
{
return PID(seq1, seq2, start, end, true, false);
}
* @param ungappedOnly
* - if true - only count PID over ungapped columns
* @return
+ * @deprecated use PIDModel.computePID()
*/
+ @Deprecated
public final static float PID(String seq1, String seq2, int start,
int end, boolean wcGaps, boolean ungappedOnly)
{
}
/**
- * DOCUMENT ME!
+ * Answers true if the supplied character is a recognised gap character, else
+ * false. Currently hard-coded to recognise '-', '-' or ' ' (hyphen / dot /
+ * space).
*
* @param c
- * DOCUMENT ME!
*
- * @return DOCUMENT ME!
+ * @return
*/
public static final boolean isGap(char c)
{
- return (c == '-' || c == '.' || c == ' ') ? true : false;
+ return (c == GAP_DASH || c == GAP_DOT || c == GAP_SPACE) ? true : false;
+ }
+
+ /**
+ * Overloaded method signature to test whether a single sequence is nucleotide
+ * (that is, more than 85% CGTA)
+ *
+ * @param seq
+ * @return
+ */
+ public static final boolean isNucleotide(SequenceI seq)
+ {
+ return isNucleotide(new SequenceI[] { seq });
}
+ /**
+ * Answers true if more than 85% of the sequence residues (ignoring gaps) are
+ * A, G, C, T or U, else false. This is just a heuristic guess and may give a
+ * wrong answer (as AGCT are also amino acid codes).
+ *
+ * @param seqs
+ * @return
+ */
public static final boolean isNucleotide(SequenceI[] seqs)
{
- int i = 0, iSize = seqs.length, j, jSize;
- float nt = 0, aa = 0;
- char c;
- while (i < iSize)
+ if (seqs == null)
{
- jSize = seqs[i].getLength();
- for (j = 0; j < jSize; j++)
+ return false;
+ }
+ char[][] letters = new char[seqs.length][];
+ for (int i = 0; i < seqs.length; i++)
+ {
+ if (seqs[i] != null)
{
- c = seqs[i].getCharAt(j);
- if ('a' <= c && c <= 'z')
+ char[] sequence = seqs[i].getSequence();
+ if (sequence != null)
{
- c -= ('a' - 'A');
+ letters[i] = sequence;
}
+ }
+ }
+
+ return areNucleotide(letters);
+ }
- if (c == 'A' || c == 'G' || c == 'C' || c == 'T' || c == 'U')
+ /**
+ * Answers true if more than 85% of the sequence residues (ignoring gaps) are
+ * A, G, C, T or U, else false. This is just a heuristic guess and may give a
+ * wrong answer (as AGCT are also amino acid codes).
+ *
+ * @param letters
+ * @return
+ */
+ static final boolean areNucleotide(char[][] letters)
+ {
+ int ntCount = 0;
+ int aaCount = 0;
+ for (char[] seq : letters)
+ {
+ if (seq == null)
+ {
+ continue;
+ }
+ // TODO could possibly make an informed guess just from the first sequence
+ // to save a lengthy calculation
+ for (char c : seq)
+ {
+ if (isNucleotide(c))
{
- nt++;
+ ntCount++;
}
- else if (!jalview.util.Comparison.isGap(seqs[i].getCharAt(j)))
+ else if (!isGap(c))
{
- aa++;
+ aaCount++;
}
}
- i++;
}
- if ((nt / (nt + aa)) > 0.85f)
+ /*
+ * Check for nucleotide count > 85% of total count (in a form that evades
+ * int / float conversion or divide by zero).
+ */
+ if (ntCount * 100 > EIGHTY_FIVE * (ntCount + aaCount))
{
return true;
}
}
}
+
+ /**
+ * Answers true if the character is one of aAcCgGtTuU
+ *
+ * @param c
+ * @return
+ */
+ public static boolean isNucleotide(char c)
+ {
+ if ('a' <= c && c <= 'z')
+ {
+ c -= TO_UPPER_CASE;
+ }
+
+ switch (c)
+ {
+ case 'A':
+ case 'C':
+ case 'G':
+ case 'T':
+ case 'U':
+ return true;
+ }
+ return false;
+ }
+
+ /**
+ * Answers true if every character in the string is one of aAcCgGtTuU, or
+ * (optionally) a gap character (dot, dash, space), else false
+ *
+ * @param s
+ * @param allowGaps
+ * @return
+ */
+ public static boolean isNucleotideSequence(String s, boolean allowGaps)
+ {
+ if (s == null)
+ {
+ return false;
+ }
+ for (int i = 0; i < s.length(); i++)
+ {
+ char c = s.charAt(i);
+ if (!isNucleotide(c))
+ {
+ if (!allowGaps || !isGap(c))
+ {
+ return false;
+ }
+ }
+ }
+ return true;
+ }
+
+ /**
+ * Convenience overload of isNucleotide
+ *
+ * @param seqs
+ * @return
+ */
+ public static boolean isNucleotide(SequenceI[][] seqs)
+ {
+ if (seqs == null)
+ {
+ return false;
+ }
+ List<SequenceI> flattened = new ArrayList<SequenceI>();
+ for (SequenceI[] ss : seqs)
+ {
+ for (SequenceI s : ss)
+ {
+ flattened.add(s);
+ }
+ }
+ final SequenceI[] oneDArray = flattened
+ .toArray(new SequenceI[flattened.size()]);
+ return isNucleotide(oneDArray);
+ }
+
+ /**
+ * Compares two residues either case sensitively or case insensitively
+ * depending on the caseSensitive flag
+ *
+ * @param c1
+ * first char
+ * @param c2
+ * second char to compare with
+ * @param caseSensitive
+ * if true comparison will be case sensitive otherwise its not
+ * @return
+ */
+ public static boolean isSameResidue(char c1, char c2,
+ boolean caseSensitive)
+ {
+ if (caseSensitive)
+ {
+ return (c1 == c2);
+ }
+ else
+ {
+ return Character.toUpperCase(c1) == Character.toUpperCase(c2);
+ }
+ }
}
git://source.jalview.org / jalview.git / blobdiff