-package jalview.datamodel;\r
-\r
-import java.util.Enumeration;\r
-import java.util.Vector;\r
-\r
-import jalview.util.MapList;\r
-\r
-/**\r
- * Stores mapping between the columns of a protein alignment and a DNA alignment\r
- * and a list of individual codon to amino acid mappings between sequences.\r
- */\r
-\r
-public class AlignedCodonFrame\r
-{\r
- /**\r
- * array of nucleotide positions for aligned codons at column of aligned proteins.\r
- */\r
- public int[][] codons = null;\r
- /**\r
- * width of protein sequence alignement\r
- * implicit assertion that codons.length >= aaWidth\r
- */\r
- public int aaWidth=0;\r
- /**\r
- * initialise codon frame with a nominal alignment width\r
- * @param aWidth\r
- */\r
- public AlignedCodonFrame(int aWidth)\r
- {\r
- if (aWidth<=0)\r
- {\r
- codons=null;\r
- return;\r
- }\r
- codons = new int[aWidth][];\r
- for (int res = 0; res < aWidth; res++)\r
- codons[res] = null;\r
- }\r
-\r
- /**\r
- * ensure that codons array is at least as wide as aslen residues\r
- * @param aslen\r
- * @return (possibly newly expanded) codon array\r
- */\r
- public int[][] checkCodonFrameWidth(int aslen)\r
- {\r
- if (codons.length <= aslen + 1)\r
- {\r
- // probably never have to do this ?\r
- int[][] c = new int[codons.length + 10][];\r
- for (int i = 0; i < codons.length; i++)\r
- {\r
- c[i] = codons[i];\r
- codons[i] = null;\r
- }\r
- codons = c;\r
- }\r
- return codons;\r
- }\r
- /**\r
- * @return width of aligned translated amino acid residues \r
- */\r
- public int getaaWidth()\r
- {\r
- return aaWidth;\r
- }\r
- /**\r
- * TODO: not an ideal solution - we reference the aligned amino acid sequences in order to make insertions on them\r
- * Better would be dnaAlignment and aaAlignment reference....\r
- */\r
- Vector a_aaSeqs=new Vector();\r
- /**\r
- * increase aaWidth by one and insert a new aligned codon position space at aspos.\r
- * @param aspos\r
- */\r
- public void insertAAGap(int aspos, char gapCharacter)\r
- { \r
- // this aa appears before the aligned codons at aspos - so shift them in each pair of mapped sequences\r
- aaWidth++;\r
- if (a_aaSeqs!=null)\r
- {\r
- // we actually have to modify the aligned sequences here, so use the a_aaSeqs vector\r
- Enumeration sq = a_aaSeqs.elements();\r
- while (sq.hasMoreElements())\r
- {\r
- ((SequenceI) sq.nextElement()).insertCharAt(aspos, gapCharacter);\r
- }\r
- }\r
- checkCodonFrameWidth(aspos);\r
- if (aspos<aaWidth)\r
- {\r
- aaWidth++;\r
- System.arraycopy(codons, aspos, codons, aspos+1, aaWidth-aspos);\r
- codons[aspos]=null; // clear so new codon position can be marked.\r
- }\r
- }\r
-\r
- public void setAaWidth(int aapos)\r
- {\r
- aaWidth = aapos;\r
- }\r
- /**\r
- * tied array of na Sequence objects.\r
- */\r
- SequenceI[] dnaSeqs=null;\r
- /**\r
- * tied array of Mappings to protein sequence Objects and \r
- SequenceI[] aaSeqs=null;\r
- * MapLists where eac maps from the corresponding dnaSeqs element to corresponding aaSeqs element \r
- */\r
- Mapping[] dnaToProt=null;\r
- /**\r
- * add a mapping between the dataset sequences for the associated dna and protein sequence objects\r
- * @param dnaseq\r
- * @param aaseq\r
- * @param map\r
- */\r
- public void addMap(SequenceI dnaseq, SequenceI aaseq, MapList map)\r
- { \r
- int nlen=1;\r
- if (dnaSeqs!=null)\r
- {\r
- nlen = dnaSeqs.length+1;\r
- }\r
- SequenceI[] ndna = new SequenceI[nlen];\r
- Mapping[] ndtp = new Mapping[nlen];\r
- if (dnaSeqs!=null)\r
- {\r
- System.arraycopy(dnaSeqs,0,ndna, 0, dnaSeqs.length);\r
- System.arraycopy(dnaToProt,0,ndtp, 0, dnaSeqs.length);\r
- }\r
- dnaSeqs = ndna;\r
- dnaToProt = ndtp;\r
- nlen--;\r
- dnaSeqs[nlen] = (dnaseq.getDatasetSequence()==null) ? dnaseq : dnaseq.getDatasetSequence();\r
- Mapping mp = new Mapping(map);\r
- mp.to = (aaseq.getDatasetSequence()==null) ? aaseq : aaseq.getDatasetSequence();\r
- a_aaSeqs.addElement(aaseq);\r
- dnaToProt[nlen] = mp;\r
- }\r
-\r
-\r
- public SequenceI[] getdnaSeqs()\r
- {\r
- return dnaSeqs;\r
- }\r
- public SequenceI[] getAaSeqs()\r
- {\r
- if (dnaToProt==null)\r
- return null;\r
- SequenceI[] sqs = new SequenceI[dnaToProt.length];\r
- for (int sz=0; sz<dnaToProt.length; sz++)\r
- {\r
- sqs[sz] = dnaToProt[sz].to;\r
- }\r
- return sqs;\r
- }\r
- public MapList[] getdnaToProt()\r
- {\r
- if (dnaToProt==null)\r
- return null;\r
- MapList[] sqs = new MapList[dnaToProt.length];\r
- for (int sz=0; sz<dnaToProt.length; sz++)\r
- {\r
- sqs[sz] = dnaToProt[sz].map;\r
- }\r
- return sqs;\r
- }\r
- public Mapping[] getProtMappings()\r
- {\r
- return dnaToProt;\r
- }\r
- /**\r
- * \r
- * @param sequenceRef\r
- * @return null or corresponding aaSeq entry for dnaSeq entry\r
- */\r
- public SequenceI getAaForDnaSeq(SequenceI dnaSeqRef)\r
- {\r
- if (dnaSeqs==null)\r
- {\r
- return null;\r
- }\r
- SequenceI dnads = dnaSeqRef.getDatasetSequence();\r
- for (int ds=0;ds<dnaSeqs.length; ds++)\r
- {\r
- if (dnaSeqs[ds]==dnaSeqRef || dnaSeqs[ds]==dnads)\r
- return dnaToProt[ds].to;\r
- }\r
- return null;\r
- }\r
- /**\r
- * \r
- * @param sequenceRef\r
- * @return null or corresponding aaSeq entry for dnaSeq entry\r
- */\r
- public SequenceI getDnaForAaSeq(SequenceI aaSeqRef)\r
- {\r
- if (dnaToProt==null)\r
- {\r
- return null;\r
- }\r
- SequenceI aads = aaSeqRef.getDatasetSequence();\r
- for (int as=0;as<dnaToProt.length; as++)\r
- {\r
- if (dnaToProt[as].to==aaSeqRef || dnaToProt[as].to==aads)\r
- return dnaSeqs[as];\r
- }\r
- return null;\r
- }\r
- /**\r
- * test to see if codon frame involves seq in any way\r
- * @param seq a nucleotide or protein sequence\r
- * @return true if a mapping exists to or from this sequence to any translated sequence\r
- */\r
- public boolean involvesSequence(SequenceI seq)\r
- {\r
- return getAaForDnaSeq(seq)!=null || getDnaForAaSeq(seq)!=null;\r
- }\r
- /**\r
- * Add search results for regions in other sequences that translate or are translated from a particular position in seq \r
- * @param seq\r
- * @param index position in seq\r
- * @param results where highlighted regions go\r
- */\r
- public void markMappedRegion(SequenceI seq, int index, SearchResults results)\r
- {\r
- int[] codon;\r
- SequenceI ds = seq.getDatasetSequence();\r
- for (int mi = 0; mi<dnaToProt.length; mi++)\r
- {\r
- if (dnaSeqs[mi]==seq || dnaSeqs[mi]==ds)\r
- {\r
- // DEBUG System.err.println("dna pos "+index);\r
- codon = dnaToProt[mi].map.locateInTo(index,index);\r
- if (codon!=null)\r
- {\r
- for (int i=0; i<codon.length; i+=2)\r
- {\r
- results.addResult(dnaToProt[mi].to, codon[i], codon[i+1]);\r
- }\r
- }\r
- } else\r
- if (dnaToProt[mi].to==seq || dnaToProt[mi].to==ds)\r
- {\r
- // DEBUG System.err.println("aa pos "+index);\r
- {\r
- codon = dnaToProt[mi].map.locateInFrom(index, index);\r
- if (codon!=null)\r
- {\r
- for (int i=0; i<codon.length; i+=2)\r
- {\r
- results.addResult(dnaSeqs[mi], codon[i], codon[i+1]);\r
- }\r
- }\r
- }\r
- }\r
- }\r
- }\r
-}\r
+/*
+ * Jalview - A Sequence Alignment Editor and Viewer (Version 2.9.0b2)
+ * Copyright (C) 2015 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 <http://www.gnu.org/licenses/>.
+ * The Jalview Authors are detailed in the 'AUTHORS' file.
+ */
+package jalview.datamodel;
+
+import jalview.util.MapList;
+import jalview.util.MappingUtils;
+
+import java.util.ArrayList;
+import java.util.List;
+
+/**
+ * Stores mapping between the columns of a protein alignment and a DNA alignment
+ * and a list of individual codon to amino acid mappings between sequences.
+ */
+public class AlignedCodonFrame
+{
+
+ /**
+ * tied array of na Sequence objects.
+ */
+ private SequenceI[] dnaSeqs = null;
+
+ /**
+ * tied array of Mappings to protein sequence Objects and SequenceI[]
+ * aaSeqs=null; MapLists where each maps from the corresponding dnaSeqs
+ * element to corresponding aaSeqs element
+ */
+ private Mapping[] dnaToProt = null;
+
+ /**
+ * Constructor
+ */
+ public AlignedCodonFrame()
+ {
+ }
+
+ /**
+ * Adds a mapping between the dataset sequences for the associated dna and
+ * protein sequence objects
+ *
+ * @param dnaseq
+ * @param aaseq
+ * @param map
+ */
+ public void addMap(SequenceI dnaseq, SequenceI aaseq, MapList map)
+ {
+ int nlen = 1;
+ if (dnaSeqs != null)
+ {
+ nlen = dnaSeqs.length + 1;
+ }
+ SequenceI[] ndna = new SequenceI[nlen];
+ Mapping[] ndtp = new Mapping[nlen];
+ if (dnaSeqs != null)
+ {
+ System.arraycopy(dnaSeqs, 0, ndna, 0, dnaSeqs.length);
+ System.arraycopy(dnaToProt, 0, ndtp, 0, dnaSeqs.length);
+ }
+ dnaSeqs = ndna;
+ dnaToProt = ndtp;
+ nlen--;
+ dnaSeqs[nlen] = (dnaseq.getDatasetSequence() == null) ? dnaseq : dnaseq
+ .getDatasetSequence();
+ Mapping mp = new Mapping(map);
+ // JBPNote DEBUG! THIS !
+ // dnaseq.transferAnnotation(aaseq, mp);
+ // aaseq.transferAnnotation(dnaseq, new Mapping(map.getInverse()));
+ mp.to = (aaseq.getDatasetSequence() == null) ? aaseq : aaseq
+ .getDatasetSequence();
+ dnaToProt[nlen] = mp;
+ }
+
+ public SequenceI[] getdnaSeqs()
+ {
+ return dnaSeqs;
+ }
+
+ public SequenceI[] getAaSeqs()
+ {
+ if (dnaToProt == null)
+ {
+ return null;
+ }
+ SequenceI[] sqs = new SequenceI[dnaToProt.length];
+ for (int sz = 0; sz < dnaToProt.length; sz++)
+ {
+ sqs[sz] = dnaToProt[sz].to;
+ }
+ return sqs;
+ }
+
+ public MapList[] getdnaToProt()
+ {
+ if (dnaToProt == null)
+ {
+ return null;
+ }
+ MapList[] sqs = new MapList[dnaToProt.length];
+ for (int sz = 0; sz < dnaToProt.length; sz++)
+ {
+ sqs[sz] = dnaToProt[sz].map;
+ }
+ return sqs;
+ }
+
+ public Mapping[] getProtMappings()
+ {
+ return dnaToProt;
+ }
+
+ /**
+ * Returns the first mapping found which is to or from the given sequence, or
+ * null.
+ *
+ * @param seq
+ * @return
+ */
+ public Mapping getMappingForSequence(SequenceI seq)
+ {
+ if (dnaSeqs == null)
+ {
+ return null;
+ }
+ SequenceI seqDs = seq.getDatasetSequence();
+ seqDs = seqDs != null ? seqDs : seq;
+
+ for (int ds = 0; ds < dnaSeqs.length; ds++)
+ {
+ if (dnaSeqs[ds] == seqDs || dnaToProt[ds].to == seqDs)
+ {
+ return dnaToProt[ds];
+ }
+ }
+ return null;
+ }
+
+ /**
+ * Return the corresponding aligned or dataset aa sequence for given dna
+ * sequence, null if not found.
+ *
+ * @param sequenceRef
+ * @return
+ */
+ public SequenceI getAaForDnaSeq(SequenceI dnaSeqRef)
+ {
+ if (dnaSeqs == null)
+ {
+ return null;
+ }
+ SequenceI dnads = dnaSeqRef.getDatasetSequence();
+ for (int ds = 0; ds < dnaSeqs.length; ds++)
+ {
+ if (dnaSeqs[ds] == dnaSeqRef || dnaSeqs[ds] == dnads)
+ {
+ return dnaToProt[ds].to;
+ }
+ }
+ return null;
+ }
+
+ /**
+ *
+ * @param sequenceRef
+ * @return null or corresponding aaSeq entry for dnaSeq entry
+ */
+ public SequenceI getDnaForAaSeq(SequenceI aaSeqRef)
+ {
+ if (dnaToProt == null)
+ {
+ return null;
+ }
+ SequenceI aads = aaSeqRef.getDatasetSequence();
+ for (int as = 0; as < dnaToProt.length; as++)
+ {
+ if (dnaToProt[as].to == aaSeqRef || dnaToProt[as].to == aads)
+ {
+ return dnaSeqs[as];
+ }
+ }
+ return null;
+ }
+
+ /**
+ * test to see if codon frame involves seq in any way
+ *
+ * @param seq
+ * a nucleotide or protein sequence
+ * @return true if a mapping exists to or from this sequence to any translated
+ * sequence
+ */
+ public boolean involvesSequence(SequenceI seq)
+ {
+ return getAaForDnaSeq(seq) != null || getDnaForAaSeq(seq) != null;
+ }
+
+ /**
+ * Add search results for regions in other sequences that translate or are
+ * translated from a particular position in seq
+ *
+ * @param seq
+ * @param index
+ * position in seq
+ * @param results
+ * where highlighted regions go
+ */
+ public void markMappedRegion(SequenceI seq, int index,
+ SearchResults results)
+ {
+ if (dnaToProt == null)
+ {
+ return;
+ }
+ int[] codon;
+ SequenceI ds = seq.getDatasetSequence();
+ for (int mi = 0; mi < dnaToProt.length; mi++)
+ {
+ if (dnaSeqs[mi] == seq || dnaSeqs[mi] == ds)
+ {
+ // DEBUG System.err.println("dna pos "+index);
+ codon = dnaToProt[mi].map.locateInTo(index, index);
+ if (codon != null)
+ {
+ for (int i = 0; i < codon.length; i += 2)
+ {
+ results.addResult(dnaToProt[mi].to, codon[i], codon[i + 1]);
+ }
+ }
+ }
+ else if (dnaToProt[mi].to == seq || dnaToProt[mi].to == ds)
+ {
+ // DEBUG System.err.println("aa pos "+index);
+ {
+ codon = dnaToProt[mi].map.locateInFrom(index, index);
+ if (codon != null)
+ {
+ for (int i = 0; i < codon.length; i += 2)
+ {
+ results.addResult(dnaSeqs[mi], codon[i], codon[i + 1]);
+ }
+ }
+ }
+ }
+ }
+ }
+
+ /**
+ * Returns the DNA codon positions (base 1) for the given position (base 1) in
+ * a mapped protein sequence, or null if no mapping is found.
+ *
+ * Intended for use in aligning cDNA to match aligned protein. Only the first
+ * mapping found is returned, so not suitable for use if multiple protein
+ * sequences are mapped to the same cDNA (but aligning cDNA as protein is
+ * ill-defined for this case anyway).
+ *
+ * @param seq
+ * the DNA dataset sequence
+ * @param aaPos
+ * residue position (base 1) in a protein sequence
+ * @return
+ */
+ public int[] getDnaPosition(SequenceI seq, int aaPos)
+ {
+ /*
+ * Adapted from markMappedRegion().
+ */
+ MapList ml = null;
+ for (int i = 0; i < dnaToProt.length; i++)
+ {
+ if (dnaSeqs[i] == seq)
+ {
+ ml = getdnaToProt()[i];
+ break;
+ }
+ }
+ return ml == null ? null : ml.locateInFrom(aaPos, aaPos);
+ }
+
+ /**
+ * Convenience method to return the first aligned sequence in the given
+ * alignment whose dataset has a mapping with the given dataset sequence.
+ *
+ * @param seq
+ *
+ * @param al
+ * @return
+ */
+ public SequenceI findAlignedSequence(SequenceI seq, AlignmentI al)
+ {
+ /*
+ * Search mapped protein ('to') sequences first.
+ */
+ if (this.dnaToProt != null)
+ {
+ for (int i = 0; i < dnaToProt.length; i++)
+ {
+ if (this.dnaSeqs[i] == seq)
+ {
+ for (SequenceI sourceAligned : al.getSequences())
+ {
+ if (this.dnaToProt[i].to == sourceAligned.getDatasetSequence())
+ {
+ return sourceAligned;
+ }
+ }
+ }
+ }
+ }
+
+ /*
+ * Then try mapped dna sequences.
+ */
+ if (this.dnaToProt != null)
+ {
+ for (int i = 0; i < dnaToProt.length; i++)
+ {
+ if (this.dnaToProt[i].to == seq)
+ {
+ for (SequenceI sourceAligned : al.getSequences())
+ {
+ if (this.dnaSeqs[i] == sourceAligned.getDatasetSequence())
+ {
+ return sourceAligned;
+ }
+ }
+ }
+ }
+ }
+
+ return null;
+ }
+
+ /**
+ * Returns the region in the 'mappedFrom' sequence's dataset that is mapped to
+ * position 'pos' (base 1) in the 'mappedTo' sequence's dataset. The region is
+ * a set of start/end position pairs.
+ *
+ * @param mappedFrom
+ * @param mappedTo
+ * @param pos
+ * @return
+ */
+ public int[] getMappedRegion(SequenceI mappedFrom, SequenceI mappedTo,
+ int pos)
+ {
+ SequenceI targetDs = mappedFrom.getDatasetSequence() == null ? mappedFrom
+ : mappedFrom.getDatasetSequence();
+ SequenceI sourceDs = mappedTo.getDatasetSequence() == null ? mappedTo
+ : mappedTo.getDatasetSequence();
+ if (targetDs == null || sourceDs == null || dnaToProt == null)
+ {
+ return null;
+ }
+ for (int mi = 0; mi < dnaToProt.length; mi++)
+ {
+ if (dnaSeqs[mi] == targetDs && dnaToProt[mi].to == sourceDs)
+ {
+ int[] codon = dnaToProt[mi].map.locateInFrom(pos, pos);
+ if (codon != null)
+ {
+ return codon;
+ }
+ }
+ }
+ return null;
+ }
+
+ /**
+ * Returns the DNA codon for the given position (base 1) in a mapped protein
+ * sequence, or null if no mapping is found.
+ *
+ * @param protein
+ * the peptide dataset sequence
+ * @param aaPos
+ * residue position (base 1) in the peptide sequence
+ * @return
+ */
+ public char[] getMappedCodon(SequenceI protein, int aaPos)
+ {
+ if (dnaToProt == null)
+ {
+ return null;
+ }
+ MapList ml = null;
+ SequenceI dnaSeq = null;
+ for (int i = 0; i < dnaToProt.length; i++)
+ {
+ if (dnaToProt[i].to == protein)
+ {
+ ml = getdnaToProt()[i];
+ dnaSeq = dnaSeqs[i];
+ break;
+ }
+ }
+ if (ml == null)
+ {
+ return null;
+ }
+ int[] codonPos = ml.locateInFrom(aaPos, aaPos);
+ if (codonPos == null)
+ {
+ return null;
+ }
+
+ /*
+ * Read off the mapped nucleotides (converting to position base 0)
+ */
+ codonPos = MappingUtils.flattenRanges(codonPos);
+ char[] dna = dnaSeq.getSequence();
+ int start = dnaSeq.getStart();
+ return new char[] { dna[codonPos[0] - start], dna[codonPos[1] - start],
+ dna[codonPos[2] - start] };
+ }
+
+ /**
+ * Returns any mappings found which are to (or from) the given sequence, and
+ * to distinct sequences.
+ *
+ * @param seq
+ * @return
+ */
+ public List<Mapping> getMappingsForSequence(SequenceI seq)
+ {
+ List<Mapping> result = new ArrayList<Mapping>();
+ if (dnaSeqs == null)
+ {
+ return result;
+ }
+ List<SequenceI> related = new ArrayList<SequenceI>();
+ SequenceI seqDs = seq.getDatasetSequence();
+ seqDs = seqDs != null ? seqDs : seq;
+
+ for (int ds = 0; ds < dnaSeqs.length; ds++)
+ {
+ final Mapping mapping = dnaToProt[ds];
+ if (dnaSeqs[ds] == seqDs || mapping.to == seqDs)
+ {
+ if (!related.contains(mapping.to))
+ {
+ result.add(mapping);
+ related.add(mapping.to);
+ }
+ }
+ }
+ return result;
+ }
+}
git://source.jalview.org / jalview.git / blobdiff