X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Futil%2FComparison.java;h=d4fc233b901a57835f85c8a3705bb797072fa254;hb=6a18ba9e7e4bd4deee0337dfd072b959596ca4b5;hp=e224b710262d78ad5bb1a78cbeaa0b7100f0ad1e;hpb=be32c14cd8e48fe0a207cd7030cb9cd46f894678;p=jalview.git diff --git a/src/jalview/util/Comparison.java b/src/jalview/util/Comparison.java index e224b71..d4fc233 100644 --- a/src/jalview/util/Comparison.java +++ b/src/jalview/util/Comparison.java @@ -22,6 +22,9 @@ package jalview.util; import jalview.datamodel.SequenceI; +import java.util.ArrayList; +import java.util.List; + /** * Assorted methods for analysing or comparing sequences. */ @@ -31,14 +34,15 @@ public class Comparison private static final int TO_UPPER_CASE = 'a' - 'A'; - private static final char GAP_SPACE = ' '; + public static final char GAP_SPACE = ' '; - private static final char GAP_DOT = '.'; + public static final char GAP_DOT = '.'; - private static final char GAP_DASH = '-'; + public static final char GAP_DASH = '-'; - public static final String GapChars = new String(new char[] - { GAP_SPACE, GAP_DOT, GAP_DASH }); + public static final String GapChars = new String( + new char[] + { GAP_SPACE, GAP_DOT, GAP_DASH }); /** * DOCUMENT ME! @@ -68,7 +72,8 @@ public class Comparison * 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(); @@ -94,8 +99,8 @@ public class Comparison { 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++; } @@ -109,8 +114,8 @@ public class Comparison { 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++; } @@ -132,7 +137,9 @@ public class Comparison * @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()); @@ -141,7 +148,12 @@ public class Comparison 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); } @@ -162,7 +174,9 @@ public class Comparison * @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) { @@ -246,9 +260,21 @@ public class Comparison } /** + * 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 animo acid codes). + * wrong answer (as AGCT are also amino acid codes). * * @param seqs * @return @@ -259,22 +285,26 @@ public class Comparison { return false; } + int ntCount = 0; int aaCount = 0; for (SequenceI seq : seqs) { - for (char c : seq.getSequence()) + if (seq == null) { - if ('a' <= c && c <= 'z') - { - c -= TO_UPPER_CASE; - } - - if (c == 'A' || c == 'G' || c == 'C' || c == 'T' || c == 'U') + continue; + } + // TODO could possibly make an informed guess just from the first sequence + // to save a lengthy calculation + int len = seq.getLength(); + for (int i = 0; i < len; i++) + { + char c = seq.getCharAt(i); + if (isNucleotide(c)) { ntCount++; } - else if (!Comparison.isGap(c)) + else if (!isGap(c)) { aaCount++; } @@ -295,4 +325,107 @@ 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); + } + + /** + * 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); + } + } }