X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fdatamodel%2FAlignment.java;h=ddb58c3870ffaa3ae246aeaf7d9bf67ae0348e75;hb=c39e588784c0fbb64c7dabfcae8db4faedde10e6;hp=a770264d648a04ca733eb2159ae138bc04122bea;hpb=d69ea8f1997771890b44e4b332a7ca84fe6f0893;p=jalview.git diff --git a/src/jalview/datamodel/Alignment.java b/src/jalview/datamodel/Alignment.java index a770264..ddb58c3 100755 --- a/src/jalview/datamodel/Alignment.java +++ b/src/jalview/datamodel/Alignment.java @@ -1,6 +1,6 @@ /* * Jalview - A Sequence Alignment Editor and Viewer - * Copyright (C) 2006 AM Waterhouse, J Procter, G Barton, M Clamp, S Searle + * Copyright (C) 2007 AM Waterhouse, J Procter, G Barton, M Clamp, S Searle * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -18,824 +18,684 @@ */ package jalview.datamodel; -import jalview.analysis.*; - -import jalview.util.*; - import java.util.*; +import jalview.analysis.*; + /** Data structure to hold and manipulate a multiple sequence alignment */ -public class Alignment implements AlignmentI +public class Alignment + implements AlignmentI { - protected Alignment dataset; - protected Vector sequences; - protected Vector groups = new Vector(); - protected char gapCharacter = '-'; - protected int type = NUCLEOTIDE; - public static final int PROTEIN = 0; - public static final int NUCLEOTIDE = 1; + protected Alignment dataset; + protected Vector sequences; + protected Vector groups = new Vector(); + protected char gapCharacter = '-'; + protected int type = NUCLEOTIDE; + public static final int PROTEIN = 0; + public static final int NUCLEOTIDE = 1; - /** DOCUMENT ME!! */ - public AlignmentAnnotation[] annotations; - - HiddenSequences hiddenSequences = new HiddenSequences(this); - - private void initAlignment(SequenceI[] seqs) { - int i=0; - - if( jalview.util.Comparison.isNucleotide(seqs)) - type = NUCLEOTIDE; - else - type = PROTEIN; + /** DOCUMENT ME!! */ + public AlignmentAnnotation[] annotations; - sequences = new Vector(); + HiddenSequences hiddenSequences = new HiddenSequences(this); - for (i = 0; i < seqs.length; i++) - { - sequences.addElement(seqs[i]); - } + private void initAlignment(SequenceI[] seqs) + { + int i = 0; - } - /** Make an alignment from an array of Sequences. - * - * @param sequences - */ - public Alignment(SequenceI[] seqs) + if (jalview.util.Comparison.isNucleotide(seqs)) { - initAlignment(seqs); - } - /** - * Make a new alignment from an array of SeqCigars - * @param seqs SeqCigar[] - */ - public Alignment(SeqCigar[] alseqs) { - SequenceI[] seqs = SeqCigar.createAlignmentSequences(alseqs, gapCharacter, new ColumnSelection(), null); - initAlignment(seqs); + type = NUCLEOTIDE; } - /** - * Make a new alignment from an CigarArray - * JBPNote - can only do this when compactAlignment does not contain hidden regions. - * JBPNote - must also check that compactAlignment resolves to a set of SeqCigars - or construct them appropriately. - * @param compactAlignment CigarArray - */ - public static AlignmentI createAlignment(CigarArray compactAlignment) { - throw new Error("Alignment(CigarArray) not yet implemented"); - // this(compactAlignment.refCigars); + else + { + type = PROTEIN; } - /** - * DOCUMENT ME! - * - * @return DOCUMENT ME! - */ - public Vector getSequences() + sequences = new Vector(); + + for (i = 0; i < seqs.length; i++) { - return sequences; + sequences.addElement(seqs[i]); } - public SequenceI [] getSequencesArray() + } + + /** Make an alignment from an array of Sequences. + * + * @param sequences + */ + public Alignment(SequenceI[] seqs) + { + initAlignment(seqs); + } + + /** + * Make a new alignment from an array of SeqCigars + * @param seqs SeqCigar[] + */ + public Alignment(SeqCigar[] alseqs) + { + SequenceI[] seqs = SeqCigar.createAlignmentSequences(alseqs, gapCharacter, + new ColumnSelection(), null); + initAlignment(seqs); + } + + /** + * Make a new alignment from an CigarArray + * JBPNote - can only do this when compactAlignment does not contain hidden regions. + * JBPNote - must also check that compactAlignment resolves to a set of SeqCigars - or construct them appropriately. + * @param compactAlignment CigarArray + */ + public static AlignmentI createAlignment(CigarArray compactAlignment) + { + throw new Error("Alignment(CigarArray) not yet implemented"); + // this(compactAlignment.refCigars); + } + + /** + * DOCUMENT ME! + * + * @return DOCUMENT ME! + */ + public Vector getSequences() + { + return sequences; + } + + public SequenceI[] getSequencesArray() + { + SequenceI[] reply = new SequenceI[sequences.size()]; + for (int i = 0; i < sequences.size(); i++) { - SequenceI [] reply = new SequenceI[sequences.size()]; - for(int i=0; i width) - { - width = seqs[i].getLength(); - } - } + public void destroyAlignment() + { + if(getDataset()!=null) + getDataset().destroyAlignment(); + + dataset = null; + sequences = null; + groups = null; + annotations = null; + hiddenSequences = null; + } - int startCol = -1, endCol = -1; - boolean delete = true; - for (int i = 0; i < width; i++) - { - delete = true; - - for (j = 0; j < jSize; j++) - { - if (seqs[j].getLength() > i) - { - if (!jalview.util.Comparison.isGap(seqs[j].getCharAt(i))) - { - if(delete) - endCol = i; - - delete = false; - break; - } - } - } - - if(delete && startCol==-1) - { - startCol = i; - } - - - if (!delete && startCol > -1) - { - deleteColumns(seqs, startCol, endCol); - if (shiftrecord!=null) { - shiftrecord.addShift(startCol, 1+endCol-startCol); - } - width -= (endCol - startCol); - i -= (endCol - startCol); - startCol = -1; - endCol = -1; - } - } - if (delete && startCol > -1) - { - deleteColumns(seqs, startCol, endCol); - if (shiftrecord!=null) { - shiftrecord.addShift(startCol, 1+endCol-startCol); - } - } - } + /** + * DOCUMENT ME! + * + * @param s DOCUMENT ME! + */ + public void deleteSequence(SequenceI s) + { + deleteSequence(findIndex(s)); + } - /** Removes a range of columns (start to end inclusive). - * - * @param seqs Sequences to remove columns from - * @param start Start column in the alignment - * @param end End column in the alignment - */ - public void deleteColumns(SequenceI [] seqs, int start, int end) + /** + * DOCUMENT ME! + * + * @param i DOCUMENT ME! + */ + public void deleteSequence(int i) + { + if (i > -1 && i < getHeight()) { - for(int i=0; iseqs[j].getLength()) - { - sequences.removeElement(seqs[j]); - j--; - jSize--; - } - else - { - seqs[j].setStart(newstart); - seqs[j].setSequence(seqs[j].getSequence().substring(i)); - } - } + SequenceGroup sg = (SequenceGroup) groups.elementAt(i); + + if (sg.getSequences(null).contains(s)) + { + return sg; + } } - /** - * DOCUMENT ME! - * - * @param i DOCUMENT ME! - */ - public void trimRight(int i) - { - SequenceI[] seqs = getVisibleAndRepresentedSeqs(); - int j, jSize = seqs.length; - for (j = 0; j < jSize; j++) - { - int newend = seqs[j].findPosition(i); + return null; + } - seqs[j].setEnd(newend); - if(seqs[j].getLength()>i) - seqs[j].setSequence(seqs[j].getSequence().substring(0, i + 1)); - } - } + /** + * DOCUMENT ME! + * + * @param s DOCUMENT ME! + * + * @return DOCUMENT ME! + */ + public SequenceGroup[] findAllGroups(SequenceI s) + { + Vector temp = new Vector(); - /** - * DOCUMENT ME! - * - * @param s DOCUMENT ME! - */ - public void deleteSequence(SequenceI s) + int gSize = groups.size(); + for (int i = 0; i < gSize; i++) { - for (int i = 0; i < getHeight(); i++) - { - if (getSequenceAt(i) == s) - { - deleteSequence(i); - } - } - } + SequenceGroup sg = (SequenceGroup) groups.elementAt(i); + if (sg == null || sg.getSequences(null) == null) + { + this.deleteGroup(sg); + gSize--; + continue; + } - /** - * DOCUMENT ME! - * - * @param i DOCUMENT ME! - */ - public void deleteSequence(int i) - { - sequences.removeElementAt(i); + if (sg.getSequences(null).contains(s)) + { + temp.addElement(sg); + } } + SequenceGroup[] ret = new SequenceGroup[temp.size()]; - /** */ - public SequenceGroup findGroup(SequenceI s) + for (int i = 0; i < temp.size(); i++) { - for (int i = 0; i < this.groups.size(); i++) - { - SequenceGroup sg = (SequenceGroup) groups.elementAt(i); - - if (sg.getSequences(false).contains(s)) - { - return sg; - } - } - - return null; + ret[i] = (SequenceGroup) temp.elementAt(i); } - /** - * DOCUMENT ME! - * - * @param s DOCUMENT ME! - * - * @return DOCUMENT ME! - */ - public SequenceGroup[] findAllGroups(SequenceI s) - { - Vector temp = new Vector(); + return ret; + } - int gSize = groups.size(); - for (int i = 0; i < gSize; i++) + /** */ + public void addGroup(SequenceGroup sg) + { + if (!groups.contains(sg)) + { + if (hiddenSequences.getSize() > 0) + { + int i, iSize = sg.getSize(); + for (i = 0; i < iSize; i++) { - SequenceGroup sg = (SequenceGroup) groups.elementAt(i); - if(sg==null || sg.getSequences(false)==null) - { - this.deleteGroup(sg); - gSize--; - continue; - } - - if (sg.getSequences(false).contains(s)) - { - temp.addElement(sg); - } + if (!sequences.contains(sg.getSequenceAt(i))) + { + sg.deleteSequence(sg.getSequenceAt(i), false); + iSize--; + i--; + } } - SequenceGroup[] ret = new SequenceGroup[temp.size()]; - - for (int i = 0; i < temp.size(); i++) + if (sg.getSize() < 1) { - ret[i] = (SequenceGroup) temp.elementAt(i); + return; } + } - return ret; + groups.addElement(sg); } + } + /** + * DOCUMENT ME! + */ + public void deleteAllGroups() + { + groups.removeAllElements(); + } - - /** */ - public void addGroup(SequenceGroup sg) + /** */ + public void deleteGroup(SequenceGroup g) + { + if (groups.contains(g)) { - if (!groups.contains(sg)) - { - groups.addElement(sg); - } + groups.removeElement(g); } + } - /** - * DOCUMENT ME! - */ - public void deleteAllGroups() - { - groups.removeAllElements(); + /** */ + public SequenceI findName(String name) + { + int i = 0; - int i = 0; + while (i < sequences.size()) + { + if (getSequenceAt(i).getName().equals(name)) + { + return getSequenceAt(i); + } - while (i < sequences.size()) - { - SequenceI s = getSequenceAt(i); - s.setColor(java.awt.Color.white); - i++; - } + i++; } - /** */ - public void deleteGroup(SequenceGroup g) + return null; + } + + public SequenceI[] findSequenceMatch(String name) + { + Vector matches = new Vector(); + int i = 0; + + while (i < sequences.size()) { - if (groups.contains(g)) - { - groups.removeElement(g); - } + if (getSequenceAt(i).getName().equals(name)) + { + matches.addElement(getSequenceAt(i)); + } + i++; } - /** */ - public SequenceI findName(String name) + SequenceI[] result = new SequenceI[matches.size()]; + for (i = 0; i < result.length; i++) { - int i = 0; + result[i] = (SequenceI) matches.elementAt(i); + } - while (i < sequences.size()) - { - if (getSequenceAt(i).getName().equals(name)) - { - return getSequenceAt(i); - } + return result; - i++; - } + } - return null; - } + /** */ + public int findIndex(SequenceI s) + { + int i = 0; - public SequenceI [] findSequenceMatch(String name) + while (i < sequences.size()) { - Vector matches = new Vector(); - int i = 0; - - while (i < sequences.size()) + if (s == getSequenceAt(i)) { - if (getSequenceAt(i).getName().equals(name)) - { - matches.addElement(getSequenceAt(i)); - } - i++; + return i; } - SequenceI [] result = new SequenceI[matches.size()]; - for(i=0; i maxLength) + { + maxLength = getSequenceAt(i).getLength(); + } } - /** - * DOCUMENT ME! - * - * @return DOCUMENT ME! - */ - public int getWidth() - { - int maxLength = -1; - - for (int i = 0; i < sequences.size(); i++) - { - if (getSequenceAt(i).getLength() > maxLength) - { - maxLength = getSequenceAt(i).getLength(); - } - } + return maxLength; + } - return maxLength; - } + /** + * DOCUMENT ME! + * + * @param gc DOCUMENT ME! + */ + public void setGapCharacter(char gc) + { + gapCharacter = gc; - /** - * DOCUMENT ME! - * - * @return DOCUMENT ME! - */ - public int getMaxIdLength() + for (int i = 0; i < sequences.size(); i++) { - int max = 0; - int i = 0; - - while (i < sequences.size()) - { - SequenceI seq = getSequenceAt(i); - String tmp = seq.getName() + "/" + seq.getStart() + "-" + - seq.getEnd(); + Sequence seq = (Sequence) sequences.elementAt(i); + seq.setSequence(seq.getSequenceAsString() + .replace('.', gc) + .replace('-', gc) + .replace(' ', gc) + ); + } + } - if (tmp.length() > max) - { - max = tmp.length(); - } + /** + * DOCUMENT ME! + * + * @return DOCUMENT ME! + */ + public char getGapCharacter() + { + return gapCharacter; + } - i++; - } + /** + * DOCUMENT ME! + * + * @return DOCUMENT ME! + */ + public boolean isAligned() + { + int width = getWidth(); - return max; + for (int i = 0; i < sequences.size(); i++) + { + if (getSequenceAt(i).getLength() != width) + { + return false; + } } - /** - * DOCUMENT ME! - * - * @param gc DOCUMENT ME! - */ - public void setGapCharacter(char gc) - { - gapCharacter = gc; + return true; + } - for (int i = 0; i < sequences.size(); i++) - { - Sequence seq = (Sequence) sequences.elementAt(i); - seq.setSequence( seq.getSequence().replace('.', gc) ); - seq.setSequence( seq.getSequence().replace('-', gc) ); - seq.setSequence( seq.getSequence().replace(' ', gc) ); - } - } + /** + * DOCUMENT ME! + * + * @param aa DOCUMENT ME! + */ + public void deleteAnnotation(AlignmentAnnotation aa) + { + int aSize = 1; - /** - * DOCUMENT ME! - * - * @return DOCUMENT ME! - */ - public char getGapCharacter() + if (annotations != null) { - return gapCharacter; + aSize = annotations.length; } - - /** - * DOCUMENT ME! - * - * @return DOCUMENT ME! - */ - public boolean isAligned() + if (aSize < 1) { - int width = getWidth(); + return; + } - for (int i = 0; i < sequences.size(); i++) - { - if (getSequenceAt(i).getLength() != width) - { - return false; - } - } + AlignmentAnnotation[] temp = new AlignmentAnnotation[aSize - 1]; - return true; - } + int tIndex = 0; - /** - * DOCUMENT ME! - * - * @param aa DOCUMENT ME! - */ - public void deleteAnnotation(AlignmentAnnotation aa) + for (int i = 0; i < aSize; i++) { - int aSize = 1; + if (annotations[i] == aa) + { + continue; + } - if (annotations != null) - { - aSize = annotations.length; - } + temp[tIndex] = annotations[i]; + tIndex++; + } - AlignmentAnnotation[] temp = new AlignmentAnnotation[aSize - 1]; + annotations = temp; + } - int tIndex = 0; + /** + * DOCUMENT ME! + * + * @param aa DOCUMENT ME! + */ + public void addAnnotation(AlignmentAnnotation aa) + { + int aSize = 1; + if (annotations != null) + { + aSize = annotations.length + 1; + } - for (int i = 0; i < aSize; i++) - { - if (annotations[i] == aa) - { - continue; - } + AlignmentAnnotation[] temp = new AlignmentAnnotation[aSize]; - temp[tIndex] = annotations[i]; - tIndex++; - } - - annotations = temp; - } + temp[aSize - 1] = aa; + int i = 0; - public void adjustSequenceAnnotations() + if (aSize > 1) { - if(annotations!=null) + for (i = 0; i < (aSize - 1); i++) { - for (int a = 0; a < annotations.length; a++) - { - if (annotations[a].sequenceRef != null) - { - annotations[a].adjustForAlignment(); - } - } + temp[i] = annotations[i]; } } - /** - * DOCUMENT ME! - * - * @param aa DOCUMENT ME! - */ - public void addAnnotation(AlignmentAnnotation aa) - { - int aSize = 1; - if (annotations != null) - { - aSize = annotations.length + 1; - } - - AlignmentAnnotation[] temp = new AlignmentAnnotation[aSize]; - - temp[aSize-1] = aa; - - int i = 0; - - if (aSize > 1) - { - for (i = 0; i < (aSize-1); i++) - { - temp[i] = annotations[i]; - } - } - - annotations = temp; - } + annotations = temp; + } - public void setAnnotationIndex(AlignmentAnnotation aa, int index) + public void setAnnotationIndex(AlignmentAnnotation aa, int index) + { + if (aa == null || annotations == null || annotations.length - 1 < index) { - if(aa==null || annotations==null || annotations.length-1 maxLength; j--) - { - if (j > maxLength && !jalview.util.Comparison.isGap( - current.getCharAt(j))) - { - maxLength = j; - break; - } - } - } + public boolean padGaps() + { + boolean modified = false; - maxLength++; + //Remove excess gaps from the end of alignment + int maxLength = -1; - for (int i = 0; i < sequences.size(); - i++) + SequenceI current; + for (int i = 0; i < sequences.size(); i++) + { + current = getSequenceAt(i); + for (int j = current.getLength(); j > maxLength; j--) { - current = getSequenceAt(i); - - if (current.getLength() < maxLength) + if (j > maxLength && !jalview.util.Comparison.isGap( + current.getCharAt(j))) { - current.insertCharAt(maxLength - 1, gapCharacter); - modified=true; - } - else if(current.getLength() > maxLength) - { - current.deleteChars(maxLength, current.getLength()); + maxLength = j; + break; } } - return modified; } - public HiddenSequences getHiddenSequences() - { - return hiddenSequences; - } - SequenceI [] getVisibleAndRepresentedSeqs() + maxLength++; + + int cLength; + for (int i = 0; i < sequences.size(); + i++) { - if(hiddenSequences==null || hiddenSequences.getSize()<1) - return getSequencesArray(); + current = getSequenceAt(i); + cLength = current.getLength(); - Vector seqs = new Vector(); - SequenceI seq; - SequenceGroup hidden; - for (int i = 0; i < sequences.size(); i++) + if (cLength < maxLength) { - seq = (SequenceI) sequences.elementAt(i); - seqs.addElement(seq); - hidden = seq.getHiddenSequences(); - if(hidden!=null) - { - for(int j=0; j maxLength) + { + current.deleteChars(maxLength, current.getLength()); } - SequenceI [] result = new SequenceI[seqs.size()]; - for(int i=0; i