X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Futil%2FComparison.java;h=0beb45bcefca1ad9319236dac9d93bd18af68636;hb=19f8ee5562523508666b4472ad624b9e549a3cea;hp=9cb2430d3546721c2500d30f7bf6ba28738acf64;hpb=506d60f0e188723ddc91c26824b41ac7034df3fe;p=jalview.git
diff --git a/src/jalview/util/Comparison.java b/src/jalview/util/Comparison.java
old mode 100755
new mode 100644
index 9cb2430..0beb45b
--- a/src/jalview/util/Comparison.java
+++ b/src/jalview/util/Comparison.java
@@ -1,43 +1,55 @@
/*
- * Jalview - A Sequence Alignment Editor and Viewer (Version 2.4)
- * Copyright (C) 2008 AM Waterhouse, J Procter, G Barton, M Clamp, S Searle
+ * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
+ * Copyright (C) $$Year-Rel$$ The Jalview Authors
*
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version 2
- * of the License, or (at your option) any later version.
+ * This file is part of Jalview.
*
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
+ * Jalview is free software: you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation, either version 3
+ * of the License, or (at your option) any later version.
+ *
+ * Jalview is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
+ * PURPOSE. See the GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA
+ * along with Jalview. If not, see .
+ * The Jalview Authors are detailed in the 'AUTHORS' file.
*/
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';
+
+ private static final char GAP_SPACE = ' ';
+
+ private static final char GAP_DOT = '.';
+
+ private static final char GAP_DASH = '-';
+
+ public static final String GapChars = new String(new char[] { GAP_SPACE,
+ GAP_DOT, GAP_DASH });
/**
* DOCUMENT ME!
*
* @param ii
- * DOCUMENT ME!
+ * DOCUMENT ME!
* @param jj
- * DOCUMENT ME!
+ * DOCUMENT ME!
*
* @return DOCUMENT ME!
*/
@@ -50,13 +62,13 @@ public class Comparison
* this was supposed to be an ungapped pid calculation
*
* @param ii
- * SequenceI
+ * SequenceI
* @param jj
- * SequenceI
+ * SequenceI
* @param start
- * int
+ * int
* @param end
- * int
+ * int
* @return float
*/
public static float compare(SequenceI ii, SequenceI jj, int start, int end)
@@ -67,12 +79,12 @@ public class Comparison
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--;
}
@@ -119,9 +131,9 @@ public class Comparison
* this is a gapped PID calculation
*
* @param s1
- * SequenceI
+ * SequenceI
* @param s2
- * SequenceI
+ * SequenceI
* @return float
*/
public final static float PID(String seq1, String seq2)
@@ -134,7 +146,29 @@ public class Comparison
// Another pid with region specification
public final static float PID(String seq1, String seq2, int start, int end)
{
+ return PID(seq1, seq2, start, end, true, false);
+ }
+ /**
+ * Calculate percent identity for a pair of sequences over a particular range,
+ * with different options for ignoring gaps.
+ *
+ * @param seq1
+ * @param seq2
+ * @param start
+ * - position in seqs
+ * @param end
+ * - position in seqs
+ * @param wcGaps
+ * - if true - gaps match any character, if false, do not match
+ * anything
+ * @param ungappedOnly
+ * - if true - only count PID over ungapped columns
+ * @return
+ */
+ public final static float PID(String seq1, String seq2, int start,
+ int end, boolean wcGaps, boolean ungappedOnly)
+ {
int s1len = seq1.length();
int s2len = seq2.length();
@@ -150,16 +184,16 @@ public class Comparison
start = len - 1; // we just use a single residue for the difference
}
- int bad = 0;
+ int elen = len - start, bad = 0;
char chr1;
char chr2;
-
+ boolean agap;
for (int i = start; i < len; i++)
{
chr1 = seq1.charAt(i);
chr2 = seq2.charAt(i);
-
+ agap = isGap(chr1) || isGap(chr2);
if ('a' <= chr1 && chr1 <= 'z')
{
// TO UPPERCASE !!!
@@ -173,57 +207,127 @@ public class Comparison
chr2 -= caseShift;
}
- if (chr1 != chr2 && !isGap(chr1) && !isGap(chr2))
+ if (chr1 != chr2)
{
- bad++;
+ if (agap)
+ {
+ if (ungappedOnly)
+ {
+ elen--;
+ }
+ else if (!wcGaps)
+ {
+ bad++;
+ }
+ }
+ else
+ {
+ bad++;
+ }
}
- }
- return ((float) 100 * (len - bad)) / len;
+ }
+ if (elen < 1)
+ {
+ return 0f;
+ }
+ return ((float) 100 * (elen - bad)) / elen;
}
/**
- * 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)
+ {
+ return false;
+ }
+ char[][] letters = new char[seqs.length][];
+ for (int i = 0; i < seqs.length; i++)
{
- jSize = seqs[i].getLength();
- for (j = 0; j < jSize; j++)
+ 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;
}
+ }
+ }
- if (c == 'A' || c == 'G' || c == 'C' || c == 'T' || c == 'U')
+ return areNucleotide(letters);
+ }
+
+ /**
+ * 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;
}
@@ -233,4 +337,82 @@ public class Comparison
}
}
+
+ /**
+ * 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 flattened = new ArrayList();
+ for (SequenceI[] ss : seqs)
+ {
+ for (SequenceI s : ss)
+ {
+ flattened.add(s);
+ }
+ }
+ final SequenceI[] oneDArray = flattened.toArray(new SequenceI[flattened
+ .size()]);
+ return isNucleotide(oneDArray);
+ }
}