X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Futil%2FComparison.java;h=c491be46de07a9f3612971dac2080b8d34eaa2cb;hb=17e77c3f2949a0729322b4a8d907f3f34b6a9914;hp=9b2edba33ea4f23190219f69b530c2115f03227a;hpb=ab43013b7e357b84b4abade0dba949668dfb2a0e;p=jalview.git diff --git a/src/jalview/util/Comparison.java b/src/jalview/util/Comparison.java index 9b2edba..c491be4 100644 --- a/src/jalview/util/Comparison.java +++ b/src/jalview/util/Comparison.java @@ -1,6 +1,6 @@ /* - * Jalview - A Sequence Alignment Editor and Viewer (Version 2.8.2b1) - * Copyright (C) 2014 The Jalview Authors + * Jalview - A Sequence Alignment Editor and Viewer (Version 2.9) + * Copyright (C) 2015 The Jalview Authors * * This file is part of Jalview. * @@ -20,18 +20,28 @@ */ 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! @@ -69,12 +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--; } @@ -225,47 +235,66 @@ public class Comparison } /** - * 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; } + /** + * 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; + } + int ntCount = 0; + int aaCount = 0; + for (SequenceI seq : seqs) { - jSize = seqs[i].getLength(); - for (j = 0; j < jSize; j++) + 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.getSequence()) { - c = seqs[i].getCharAt(j); if ('a' <= c && c <= 'z') { - c -= ('a' - 'A'); + c -= TO_UPPER_CASE; } if (c == 'A' || c == 'G' || c == 'C' || c == 'T' || c == 'U') { - nt++; + ntCount++; } - else if (!jalview.util.Comparison.isGap(seqs[i].getCharAt(j))) + else if (!Comparison.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; } @@ -275,4 +304,29 @@ public class Comparison } } + + /** + * 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); + } }