package jalview.datamodel; import jalview.analysis.*; import jalview.util.*; import java.util.*; /** Data structure to hold and manipulate a multiple sequence alignment */ public class Alignment implements AlignmentI { protected Vector sequences; protected Vector groups = new Vector(); protected Vector superGroup = new Vector(); protected char gapCharacter = '-'; public AlignmentAnnotation [] annotations; public boolean featuresAdded = false; /** Make an alignment from an array of Sequences. * * @param sequences */ public Alignment(SequenceI[] seqs) { sequences = new Vector(); for (int i=0; i < seqs.length; i++) sequences.addElement(seqs[i]); getWidth(); } public Vector getSequences() { return sequences; } public SequenceI getSequenceAt(int i) { if (i < sequences.size()) { return (SequenceI)sequences.elementAt(i); } return null; } /** Adds a sequence to the alignment. Recalculates maxLength and size. * * @param snew */ public void addSequence(SequenceI snew) { sequences.addElement(snew); } public void addSequence(SequenceI[] seq) { for (int i=0; i < seq.length; i++) { addSequence(seq[i]); } } /** Adds a sequence to the alignment. Recalculates maxLength and size. * * @param snew */ public void setSequenceAt(int i,SequenceI snew) { SequenceI oldseq = getSequenceAt(i); deleteSequence(oldseq); sequences.setElementAt(snew,i); } public Vector getGroups() { return groups; } /** Sorts the sequences by sequence group size - largest to smallest. * Uses QuickSort. */ public void sortGroups() { float[] arr = new float [groups.size()]; Object[] s = new Object[groups.size()]; for (int i=0; i < groups.size(); i++) { arr[i] = ((SequenceGroup)groups.elementAt(i)).sequences.size(); s[i] = groups.elementAt(i); } QuickSort.sort(arr,s); Vector newg = new Vector(groups.size()); for (int i=groups.size()-1; i >= 0; i--) { newg.addElement(s[i]); } groups = newg; } /** Takes out columns consisting entirely of gaps (-,.," ") */ public void removeGaps() { SequenceI current; int iSize = getWidth(); for (int i=0; i < iSize; i++) { boolean delete = true; for (int j=0; j < getHeight(); j++) { current = getSequenceAt(j); if (current.getLength() > i) { /* MC Should move this to a method somewhere */ if ( !jalview.util.Comparison.isGap(current.getCharAt(i))) delete = false; } } if ( delete ) { deleteColumns(i,i); iSize--; i--; } } } /** Returns an array of Sequences containing columns * start to end (inclusive) only. * * @param start start column to fetch * @param end end column to fetch * @return Array of Sequences, ready to put into a new Alignment */ public SequenceI[] getColumns(int start, int end) { return getColumns(0,getHeight()-1,start,end); } /** Removes a range of columns (start to end inclusive). * * @param start Start column in the alignment * @param end End column in the alignment */ public void deleteColumns(int start, int end) { deleteColumns(0,getHeight()-1,start,end); } public void deleteColumns(int seq1, int seq2, int start, int end) { for (int i=0; i <= (end-start); i++) { for (int j=seq1; j <= seq2; j++) { getSequenceAt(j).deleteCharAt(start); } } } public void insertColumns(SequenceI[] seqs, int pos) { if (seqs.length == getHeight()) { for (int i=0; i < getHeight();i++) { String tmp = new String(getSequenceAt(i).getSequence()); getSequenceAt(i).setSequence(tmp.substring(0,pos) + seqs[i].getSequence() + tmp.substring(pos)); } } } public SequenceI[] getColumns(int seq1, int seq2, int start, int end) { SequenceI[] seqs = new Sequence[(seq2-seq1)+1]; for (int i=seq1; i<= seq2; i++ ) { seqs[i] = new Sequence(getSequenceAt(i).getName(), getSequenceAt(i).getSequence().substring(start,end), getSequenceAt(i).findPosition(start), getSequenceAt(i).findPosition(end)); } return seqs; } public void trimLeft(int i) { for (int j = 0;j< getHeight();j++) { SequenceI s = getSequenceAt(j); int newstart = s.findPosition(i); s.setStart(newstart); s.setSequence(s.getSequence().substring(i)); } } public void trimRight(int i) { for (int j = 0;j< getHeight();j++) { SequenceI s = getSequenceAt(j); int newend = s.findPosition(i); s.setEnd(newend); s.setSequence(s.getSequence().substring(0,i+1)); } } public void deleteSequence(SequenceI s) { for (int i=0; i < getHeight(); i++) if (getSequenceAt(i) == s) deleteSequence(i); } public void deleteSequence(int i) { sequences.removeElementAt(i); } public Vector removeRedundancy(float threshold, Vector sel) { Vector del = new Vector(); for (int i = 1; i < sel.size(); i++) { for (int j = 0; j < i; j++) { // Only do the comparison if either have not been deleted if (!del.contains( (SequenceI) sel.elementAt(i)) || !del.contains( (SequenceI) sel.elementAt(j))) { float pid = Comparison.compare( (SequenceI) sel.elementAt(j), (SequenceI) sel.elementAt(i)); if (pid >= threshold) { // Delete the shortest one if ( ( (SequenceI) sel.elementAt(j)).getSequence().length() > ( (SequenceI) sel.elementAt(i)).getSequence().length()) del.addElement(sel.elementAt(i)); else del.addElement(sel.elementAt(i)); } } } } // Now delete the sequences for (int i=0; i < del.size(); i++) deleteSequence((SequenceI)del.elementAt(i)); return del; } public void sortByPID(SequenceI s) { float scores[] = new float[getHeight()]; SequenceI seqs[] = new SequenceI[getHeight()]; for (int i = 0; i < getHeight(); i++) { scores[i] = Comparison.compare(getSequenceAt(i),s); seqs[i] = getSequenceAt(i); } QuickSort.sort(scores,0,scores.length-1,seqs); int len = 0; if (getHeight()%2 == 0) { len = getHeight()/2; } else { len = (getHeight()+1)/2; } for (int i = 0; i < len; i++) { SequenceI tmp = seqs[i]; sequences.setElementAt(seqs[getHeight()-i-1],i); sequences.setElementAt(tmp,getHeight()-i-1); } } public void sortByID() { String ids[] = new String[getHeight()]; SequenceI seqs[] = new SequenceI[getHeight()]; for (int i = 0; i < getHeight(); i++) { ids[i] = getSequenceAt(i).getName(); seqs[i] = getSequenceAt(i); } QuickSort.sort(ids,seqs); int len = 0; if (getHeight()%2 == 0) { len = getHeight()/2; } else { len = (getHeight()+1)/2; System.out.println("Sort len is odd = " + len); } for (int i = 0; i < len; i++) { System.out.println("Swapping " + seqs[i].getName() + " and " + seqs[getHeight()-i-1].getName()); SequenceI tmp = seqs[i]; sequences.setElementAt(seqs[getHeight()-i-1],i); sequences.setElementAt(tmp,getHeight()-i-1); } } /** */ public SequenceGroup findGroup(int i) { return findGroup(getSequenceAt(i)); } /** */ public SequenceGroup findGroup(SequenceI s) { for (int i = 0; i < this.groups.size();i++) { SequenceGroup sg = (SequenceGroup)groups.elementAt(i); if (sg.sequences.contains(s)) return sg; } return null; } public SequenceGroup [] findAllGroups(SequenceI s) { Vector temp = new Vector(); for (int i = 0; i < this.groups.size();i++) { SequenceGroup sg = (SequenceGroup)groups.elementAt(i); if (sg.sequences.contains(s)) temp.addElement(sg); } SequenceGroup [] ret = new SequenceGroup[temp.size()]; for(int i=0; i maxLength) maxLength = getSequenceAt(i).getLength(); } return maxLength; } public int getMaxIdLength() { int max = 0; int i = 0; while (i < sequences.size()) { SequenceI seq = getSequenceAt(i); String tmp = seq.getName() + "/" + seq.getStart() + "-" + seq.getEnd(); if (tmp.length() > max) { max = tmp.length(); } i++; } return max; } public void setGapCharacter(char gc) { gapCharacter = gc; for (int i=0; i < sequences.size(); i++) { Sequence seq = (Sequence)sequences.elementAt(i); seq.sequence = seq.sequence.replace('.', gc); seq.sequence = seq.sequence.replace('-', gc); } } public char getGapCharacter() { return gapCharacter; } public Vector getAAFrequency() { return AAFrequency.calculate(sequences, 0, getWidth()); } public boolean isAligned() { int width = getWidth(); for (int i = 0; i < sequences.size(); i++) if (getSequenceAt(i).getLength() != width) return false; return true; } public void deleteAnnotation(AlignmentAnnotation aa) { int aSize = 1; if(annotations!=null) aSize = annotations.length; AlignmentAnnotation [] temp = new AlignmentAnnotation [aSize-1]; int tIndex = 0; for (int i = 0; i < aSize; i++) { if(annotations[i]==aa) continue; temp[tIndex] = annotations[i]; tIndex++; } annotations = temp; } public void addAnnotation(AlignmentAnnotation aa) { int aSize = 1; if(annotations!=null) aSize = annotations.length+1; AlignmentAnnotation [] temp = new AlignmentAnnotation [aSize]; int i=0; if (aSize > 1) for (i = 0; i < aSize-1; i++) temp[i] = annotations[i]; temp[i] = aa; annotations = temp; } public AlignmentAnnotation[] getAlignmentAnnotation() { return annotations; } }