X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fanalysis%2FDna.java;h=8bf0c24e61fa6a125863353afcf9f70a91ea8b46;hb=042832306c7e7aa0484079f3bab1e141b2944de5;hp=800cef283ba334d9c04a6b91c94b22f8a57cea8a;hpb=56d75bc01515a2709263a5b30de447f28c4fa5c1;p=jalview.git diff --git a/src/jalview/analysis/Dna.java b/src/jalview/analysis/Dna.java index 800cef2..8bf0c24 100644 --- a/src/jalview/analysis/Dna.java +++ b/src/jalview/analysis/Dna.java @@ -44,8 +44,8 @@ import jalview.util.ShiftList; import java.util.ArrayList; import java.util.Arrays; import java.util.Comparator; +import java.util.Iterator; import java.util.List; -import java.util.Map; public class Dna { @@ -57,19 +57,23 @@ public class Dna * 'final' variables describe the inputs to the translation, which should not * be modified. */ - final private List selection; + private final List selection; - final private String[] seqstring; + private final String[] seqstring; - final private int[] contigs; + private final Iterator contigs; - final private char gapChar; + private final char gapChar; - final private AlignmentAnnotation[] annotations; + private final AlignmentAnnotation[] annotations; - final private int dnaWidth; + private final int dnaWidth; - final private AlignmentI dataset; + private final AlignmentI dataset; + + private ShiftList vismapping; + + private int[] startcontigs; /* * Working variables for the translation. @@ -92,7 +96,7 @@ public class Dna * @param viewport * @param visibleContigs */ - public Dna(AlignViewportI viewport, int[] visibleContigs) + public Dna(AlignViewportI viewport, Iterator visibleContigs) { this.selection = Arrays.asList(viewport.getSequenceSelection()); this.seqstring = viewport.getViewAsString(true); @@ -101,6 +105,45 @@ public class Dna this.annotations = viewport.getAlignment().getAlignmentAnnotation(); this.dnaWidth = viewport.getAlignment().getWidth(); this.dataset = viewport.getAlignment().getDataset(); + initContigs(); + } + + /** + * Initialise contigs used as starting point for translateCodingRegion + */ + private void initContigs() + { + vismapping = new ShiftList(); // map from viscontigs to seqstring + // intervals + + int npos = 0; + int[] lastregion = null; + ArrayList tempcontigs = new ArrayList<>(); + while (contigs.hasNext()) + { + int[] region = contigs.next(); + if (lastregion == null) + { + vismapping.addShift(npos, region[0]); + } + else + { + // hidden region + vismapping.addShift(npos, region[0] - lastregion[1] + 1); + } + lastregion = region; + tempcontigs.add(region[0]); + tempcontigs.add(region[1]); + } + + startcontigs = new int[tempcontigs.size()]; + int i = 0; + for (Integer val : tempcontigs) + { + startcontigs[i] = val; + i++; + } + tempcontigs = null; } /** @@ -134,7 +177,8 @@ public class Dna * @param ac2 * @return */ - private static int jalview_2_8_2compare(AlignedCodon ac1, AlignedCodon ac2) + private static int jalview_2_8_2compare(AlignedCodon ac1, + AlignedCodon ac2) { if (ac1 == null || ac2 == null || (ac1.equals(ac2))) { @@ -150,10 +194,11 @@ public class Dna } /** + * Translates cDNA using the specified code table * * @return */ - public AlignmentI translateCdna() + public AlignmentI translateCdna(GeneticCodeI codeTable) { AlignedCodonFrame acf = new AlignedCodonFrame(); @@ -161,11 +206,11 @@ public class Dna int s; int sSize = selection.size(); - List pepseqs = new ArrayList(); + List pepseqs = new ArrayList<>(); for (s = 0; s < sSize; s++) { SequenceI newseq = translateCodingRegion(selection.get(s), - seqstring[s], acf, pepseqs); + seqstring[s], acf, pepseqs, codeTable); if (newseq != null) { @@ -208,29 +253,30 @@ public class Dna for (int gd = 0; gd < selection.length; gd++) { SequenceI dna = selection[gd]; - DBRefEntry[] dnarefs = DBRefUtils.selectRefs(dna.getDBRefs(), + List dnarefs = DBRefUtils.selectRefs(dna.getDBRefs(), jalview.datamodel.DBRefSource.DNACODINGDBS); if (dnarefs != null) { // intersect with pep - List mappedrefs = new ArrayList(); - DBRefEntry[] refs = dna.getDBRefs(); - for (int d = 0; d < refs.length; d++) + List mappedrefs = new ArrayList<>(); + List refs = dna.getDBRefs(); + for (int d = 0, nd = refs.size(); d < nd; d++) { - if (refs[d].getMap() != null && refs[d].getMap().getMap() != null - && refs[d].getMap().getMap().getFromRatio() == 3 - && refs[d].getMap().getMap().getToRatio() == 1) + DBRefEntry ref = refs.get(d); + if (ref.getMap() != null && ref.getMap().getMap() != null + && ref.getMap().getMap().getFromRatio() == 3 + && ref.getMap().getMap().getToRatio() == 1) { - mappedrefs.add(refs[d]); // add translated protein maps + mappedrefs.add(ref); // add translated protein maps } } - dnarefs = mappedrefs.toArray(new DBRefEntry[mappedrefs.size()]); - for (int d = 0; d < dnarefs.length; d++) + dnarefs = mappedrefs;//.toArray(new DBRefEntry[mappedrefs.size()]); + for (int d = 0, nd = dnarefs.size(); d < nd; d++) { - Mapping mp = dnarefs[d].getMap(); + Mapping mp = dnarefs.get(d).getMap(); if (mp != null) { - for (int vc = 0; vc < viscontigs.length; vc += 2) + for (int vc = 0, nv = viscontigs.length; vc < nv; vc += 2) { int[] mpr = mp.locateMappedRange(viscontigs[vc], viscontigs[vc + 1]); @@ -385,33 +431,21 @@ public class Dna * @param acf * Definition of global ORF alignment reference frame * @param proteinSeqs + * @param codeTable * @return sequence ready to be added to alignment. */ protected SequenceI translateCodingRegion(SequenceI selection, String seqstring, AlignedCodonFrame acf, - List proteinSeqs) + List proteinSeqs, GeneticCodeI codeTable) { - List skip = new ArrayList(); - int skipint[] = null; - ShiftList vismapping = new ShiftList(); // map from viscontigs to seqstring - // intervals - int vc; - int[] scontigs = new int[contigs.length]; + List skip = new ArrayList<>(); + int[] skipint = null; + int npos = 0; - for (vc = 0; vc < contigs.length; vc += 2) - { - if (vc == 0) - { - vismapping.addShift(npos, contigs[vc]); - } - else - { - // hidden region - vismapping.addShift(npos, contigs[vc] - contigs[vc - 1] + 1); - } - scontigs[vc] = contigs[vc]; - scontigs[vc + 1] = contigs[vc + 1]; - } + int vc = 0; + + int[] scontigs = new int[startcontigs.length]; + System.arraycopy(startcontigs, 0, scontigs, 0, startcontigs.length); // allocate a roughly sized buffer for the protein sequence StringBuilder protein = new StringBuilder(seqstring.length() / 2); @@ -436,7 +470,7 @@ public class Dna * Filled up a reading frame... */ AlignedCodon alignedCodon = new AlignedCodon(cdp[0], cdp[1], cdp[2]); - String aa = ResidueProperties.codonTranslate(new String(codon)); + String aa = codeTable.translate(new String(codon)); rf = 0; final String gapString = String.valueOf(gapChar); if (aa == null) @@ -444,10 +478,11 @@ public class Dna aa = gapString; if (skipint == null) { - skipint = new int[] { alignedCodon.pos1, alignedCodon.pos3 /* - * cdp[0], - * cdp[2] - */}; + skipint = new int[] { alignedCodon.pos1, + alignedCodon.pos3 /* + * cdp[0], + * cdp[2] + */ }; } skipint[1] = alignedCodon.pos3; // cdp[2]; } @@ -502,8 +537,8 @@ public class Dna } if (vc + 2 < t.length) { - System.arraycopy(scontigs, vc + 2, t, vc, t.length - - vc + 2); + System.arraycopy(scontigs, vc + 2, t, vc, + t.length - vc + 2); } scontigs = t; } @@ -542,7 +577,7 @@ public class Dna skip.add(skipint); skipint = null; } - if (aa.equals("STOP")) + if (aa.equals(ResidueProperties.STOP)) { aa = STOP_ASTERIX; } @@ -596,9 +631,9 @@ public class Dna } else if (!alignedCodons[aspos].equals(alignedCodon)) { - throw new IllegalStateException("Tried to coalign " - + alignedCodons[aspos].toString() + " with " - + alignedCodon.toString()); + throw new IllegalStateException( + "Tried to coalign " + alignedCodons[aspos].toString() + + " with " + alignedCodon.toString()); } if (aspos >= aaWidth) { @@ -685,7 +720,7 @@ public class Dna */ MapList map = new MapList(scontigs, new int[] { 1, resSize }, 3, 1); - transferCodedFeatures(selection, newseq, map, null, null); + transferCodedFeatures(selection, newseq, map); /* * Construct a dataset sequence for our new peptide. @@ -754,56 +789,38 @@ public class Dna /** * Given a peptide newly translated from a dna sequence, copy over and set any - * features on the peptide from the DNA. If featureTypes is null, all features - * on the dna sequence are searched (rather than just the displayed ones), and - * similarly for featureGroups. + * features on the peptide from the DNA. * * @param dna * @param pep * @param map - * @param featureTypes - * hash whose keys are the displayed feature type strings - * @param featureGroups - * hash where keys are feature groups and values are Boolean objects - * indicating if they are displayed. */ private static void transferCodedFeatures(SequenceI dna, SequenceI pep, - MapList map, Map featureTypes, - Map featureGroups) + MapList map) { - SequenceFeature[] sfs = dna.getSequenceFeatures(); - Boolean fgstate; - DBRefEntry[] dnarefs = DBRefUtils.selectRefs(dna.getDBRefs(), - DBRefSource.DNACODINGDBS); - if (dnarefs != null) + // BH 2019.01.25 nop? +// List dnarefs = DBRefUtils.selectRefs(dna.getDBRefs(), +// DBRefSource.DNACODINGDBS); +// if (dnarefs != null) +// { +// // intersect with pep +// for (int d = 0, nd = dnarefs.size(); d < nd; d++) +// { +// Mapping mp = dnarefs.get(d).getMap(); +// if (mp != null) +// { +// } +// } +// } + for (SequenceFeature sf : dna.getFeatures().getAllFeatures()) { - // intersect with pep - for (int d = 0; d < dnarefs.length; d++) - { - Mapping mp = dnarefs[d].getMap(); - if (mp != null) - { - } - } - } - if (sfs != null) - { - for (SequenceFeature sf : sfs) - { - fgstate = (featureGroups == null) ? null : featureGroups - .get(sf.featureGroup); - if ((featureTypes == null || featureTypes.containsKey(sf.getType())) - && (fgstate == null || fgstate.booleanValue())) + if (FeatureProperties.isCodingFeature(null, sf.getType())) { - if (FeatureProperties.isCodingFeature(null, sf.getType())) + // if (map.intersectsFrom(sf[f].begin, sf[f].end)) { - // if (map.intersectsFrom(sf[f].begin, sf[f].end)) - { - } } } - } } } @@ -817,7 +834,7 @@ public class Dna public AlignmentI reverseCdna(boolean complement) { int sSize = selection.size(); - List reversed = new ArrayList(); + List reversed = new ArrayList<>(); for (int s = 0; s < sSize; s++) { SequenceI newseq = reverseSequence(selection.get(s).getName(), @@ -868,6 +885,23 @@ public class Dna } /** + * Answers the reverse complement of the input string + * + * @see #getComplement(char) + * @param s + * @return + */ + public static String reverseComplement(String s) + { + StringBuilder sb = new StringBuilder(s.length()); + for (int i = s.length() - 1; i >= 0; i--) + { + sb.append(Dna.getComplement(s.charAt(i))); + } + return sb.toString(); + } + + /** * Returns dna complement (preserving case) for aAcCgGtTuU. Ambiguity codes * are treated as on http://reverse-complement.com/. Anything else is left * unchanged. @@ -878,7 +912,8 @@ public class Dna public static char getComplement(char c) { char result = c; - switch (c) { + switch (c) + { case '-': case '.': case ' ':