/* * Jalview - A Sequence Alignment Editor and Viewer (Version 2.8.2) * Copyright (C) 2014 The Jalview Authors * * This file is part of Jalview. * * 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 Jalview. If not, see . * The Jalview Authors are detailed in the 'AUTHORS' file. */ package jalview.analysis; import jalview.datamodel.AlignedCodonFrame; import jalview.datamodel.AlignmentAnnotation; import jalview.datamodel.AlignmentI; import jalview.datamodel.SequenceI; import jalview.schemes.ResidueProperties; import jalview.util.MapList; import java.util.ArrayList; import java.util.LinkedHashMap; import java.util.List; import java.util.Map; /** * grab bag of useful alignment manipulation operations Expect these to be * refactored elsewhere at some point. * * @author jimp * */ public class AlignmentUtils { /** * Represents the 3 possible results of trying to map one alignment to * another. */ public enum MappingResult { Mapped, NotMapped, AlreadyMapped } /** * given an existing alignment, create a new alignment including all, or up to * flankSize additional symbols from each sequence's dataset sequence * * @param core * @param flankSize * @return AlignmentI */ public static AlignmentI expandContext(AlignmentI core, int flankSize) { List sq = new ArrayList(); int maxoffset = 0; for (SequenceI s : core.getSequences()) { SequenceI newSeq = s.deriveSequence(); if (newSeq.getStart() > maxoffset && newSeq.getDatasetSequence().getStart() < s.getStart()) { maxoffset = newSeq.getStart(); } sq.add(newSeq); } if (flankSize > -1) { maxoffset = flankSize; } // now add offset to create a new expanded alignment for (SequenceI s : sq) { SequenceI ds = s; while (ds.getDatasetSequence() != null) { ds = ds.getDatasetSequence(); } int s_end = s.findPosition(s.getStart() + s.getLength()); // find available flanking residues for sequence int ustream_ds = s.getStart() - ds.getStart(), dstream_ds = ds .getEnd() - s_end; // build new flanked sequence // compute gap padding to start of flanking sequence int offset = maxoffset - ustream_ds; // padding is gapChar x ( maxoffset - min(ustream_ds, flank) if (flankSize >= 0) { if (flankSize < ustream_ds) { // take up to flankSize residues offset = maxoffset - flankSize; ustream_ds = flankSize; } if (flankSize < dstream_ds) { dstream_ds = flankSize; } } char[] upstream = new String(ds.getSequence(s.getStart() - 1 - ustream_ds, s.getStart() - 1)).toLowerCase().toCharArray(); char[] downstream = new String(ds.getSequence(s_end - 1, s_end + 1 + dstream_ds)).toLowerCase().toCharArray(); char[] coreseq = s.getSequence(); char[] nseq = new char[offset + upstream.length + downstream.length + coreseq.length]; char c = core.getGapCharacter(); // TODO could lowercase the flanking regions int p = 0; for (; p < offset; p++) { nseq[p] = c; } // s.setSequence(new String(upstream).toLowerCase()+new String(coreseq) + // new String(downstream).toLowerCase()); System.arraycopy(upstream, 0, nseq, p, upstream.length); System.arraycopy(coreseq, 0, nseq, p + upstream.length, coreseq.length); System.arraycopy(downstream, 0, nseq, p + coreseq.length + upstream.length, downstream.length); s.setSequence(new String(nseq)); s.setStart(s.getStart() - ustream_ds); s.setEnd(s_end + downstream.length); } AlignmentI newAl = new jalview.datamodel.Alignment( sq.toArray(new SequenceI[0])); for (SequenceI s : sq) { if (s.getAnnotation() != null) { for (AlignmentAnnotation aa : s.getAnnotation()) { newAl.addAnnotation(aa); } } } newAl.setDataset(core.getDataset()); return newAl; } /** * Returns the index (zero-based position) of a sequence in an alignment, or * -1 if not found. * * @param al * @param seq * @return */ public static int getSequenceIndex(AlignmentI al, SequenceI seq) { int result = -1; int pos = 0; for (SequenceI alSeq : al.getSequences()) { if (alSeq == seq) { result = pos; break; } pos++; } return result; } /** * Returns a map of lists of sequences in the alignment, keyed by sequence * name. For use in mapping between different alignment views of the same * sequences. * * @see jalview.datamodel.AlignmentI#getSequencesByName() */ public static Map> getSequencesByName( AlignmentI al) { Map> theMap = new LinkedHashMap>(); for (SequenceI seq : al.getSequences()) { String name = seq.getName(); if (name != null) { List seqs = theMap.get(name); if (seqs == null) { seqs = new ArrayList(); theMap.put(name, seqs); } seqs.add(seq); } } return theMap; } /** * Build mapping of protein to cDNA alignment. Mappings are made between * sequences which have the same name and compatible lengths. Has a 3-valued * result: either Mapped (at least one sequence mapping was created), * AlreadyMapped (all possible sequence mappings already exist), or NotMapped * (no possible sequence mappings exist). * * @param proteinAlignment * @param cdnaAlignment * @return */ public static MappingResult mapProteinToCdna( final AlignmentI proteinAlignment, final AlignmentI cdnaAlignment) { boolean mappingPossible = false; boolean mappingPerformed = false; List thisSeqs = proteinAlignment.getSequences(); /* * Build a look-up of cDNA sequences by name, for matching purposes. */ Map> cdnaSeqs = cdnaAlignment .getSequencesByName(); for (SequenceI aaSeq : thisSeqs) { AlignedCodonFrame acf = new AlignedCodonFrame(); List candidates = cdnaSeqs.get(aaSeq.getName()); if (candidates == null) { /* * No cDNA sequence with matching name, so no mapping possible for this * protein sequence */ continue; } mappingPossible = true; for (SequenceI cdnaSeq : candidates) { if (!mappingExists(proteinAlignment.getCodonFrames(), aaSeq.getDatasetSequence(), cdnaSeq.getDatasetSequence())) { MapList map = mapProteinToCdna(aaSeq, cdnaSeq); if (map != null) { acf.addMap(cdnaSeq, aaSeq, map); mappingPerformed = true; } } } proteinAlignment.addCodonFrame(acf); } /* * If at least one mapping was possible but none was done, then the * alignments are already as mapped as they can be. */ if (mappingPossible && !mappingPerformed) { return MappingResult.AlreadyMapped; } else { return mappingPerformed ? MappingResult.Mapped : MappingResult.NotMapped; } } /** * Answers true if the mappings include one between the given (dataset) * sequences. */ public static boolean mappingExists(AlignedCodonFrame[] codonFrames, SequenceI aaSeq, SequenceI cdnaSeq) { if (codonFrames != null) { for (AlignedCodonFrame acf : codonFrames) { if (cdnaSeq == acf.getDnaForAaSeq(aaSeq)) { return true; } } } return false; } /** * Build a mapping (if possible) of a protein to a cDNA sequence. The cDNA * must be three times the length of the protein, possibly after ignoring * start and/or stop codons. Returns null if no mapping is determined. * * @param proteinSeqs * @param cdnaSeq * @return */ public static MapList mapProteinToCdna(SequenceI proteinSeq, SequenceI cdnaSeq) { String aaSeqString = proteinSeq.getDatasetSequence() .getSequenceAsString(); String cdnaSeqString = cdnaSeq.getDatasetSequence() .getSequenceAsString(); if (aaSeqString == null || cdnaSeqString == null) { return null; } final int mappedLength = 3 * aaSeqString.length(); int cdnaLength = cdnaSeqString.length(); int cdnaStart = 1; int cdnaEnd = cdnaLength; final int proteinStart = 1; final int proteinEnd = aaSeqString.length(); /* * If lengths don't match, try ignoring stop codon. */ if (cdnaLength != mappedLength) { for (Object stop : ResidueProperties.STOP) { if (cdnaSeqString.toUpperCase().endsWith((String) stop)) { cdnaEnd -= 3; cdnaLength -= 3; break; } } } /* * If lengths still don't match, try ignoring start codon. */ if (cdnaLength != mappedLength && cdnaSeqString.toUpperCase().startsWith( ResidueProperties.START)) { cdnaStart += 3; cdnaLength -= 3; } if (cdnaLength == mappedLength) { MapList map = new MapList(new int[] { cdnaStart, cdnaEnd }, new int[] { proteinStart, proteinEnd }, 3, 1); return map; } else { return null; } } }