X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;ds=sidebyside;f=src%2Fjalview%2Fws%2Fdatamodel%2Falphafold%2FPAEContactMatrix.java;h=ce072e06a0477b8b3245154dae6548758618a568;hb=beb2fee66dde629e3bbb7febb38d0116e1a64df2;hp=30c77d287ac150b74eb65fdbda054cd6c8f68427;hpb=7420ce36f2b43280ef610e3743960207e4c2dbe3;p=jalview.git diff --git a/src/jalview/ws/datamodel/alphafold/PAEContactMatrix.java b/src/jalview/ws/datamodel/alphafold/PAEContactMatrix.java index 30c77d2..ce072e0 100644 --- a/src/jalview/ws/datamodel/alphafold/PAEContactMatrix.java +++ b/src/jalview/ws/datamodel/alphafold/PAEContactMatrix.java @@ -1,54 +1,60 @@ package jalview.ws.datamodel.alphafold; +import java.awt.Color; +import java.io.BufferedInputStream; +import java.io.File; +import java.io.FileInputStream; +import java.io.IOException; import java.util.ArrayList; import java.util.BitSet; +import java.util.HashMap; import java.util.Iterator; import java.util.List; import java.util.Map; +import java.util.Map.Entry; + +import org.json.simple.JSONObject; import jalview.analysis.AverageDistanceEngine; import jalview.bin.Console; +import jalview.datamodel.Annotation; import jalview.datamodel.BinaryNode; import jalview.datamodel.ContactListI; import jalview.datamodel.ContactListImpl; import jalview.datamodel.ContactListProviderI; import jalview.datamodel.ContactMatrixI; +import jalview.datamodel.GroupSet; +import jalview.datamodel.GroupSetI; +import jalview.datamodel.Mapping; +import jalview.datamodel.SequenceDummy; import jalview.datamodel.SequenceI; +import jalview.io.DataSourceType; +import jalview.io.FileFormatException; +import jalview.io.FileParse; +import jalview.util.MapList; import jalview.util.MapUtils; +import jalview.ws.dbsources.EBIAlfaFold; -public class PAEContactMatrix implements ContactMatrixI +public class PAEContactMatrix extends + MappableContactMatrix implements ContactMatrixI { - SequenceI refSeq = null; - - /** - * the length that refSeq is expected to be (excluding gaps, of course) - */ - int length; int maxrow = 0, maxcol = 0; - int[] indices1, indices2; float[][] elements; float maxscore; - private void setRefSeq(SequenceI _refSeq) - { - refSeq = _refSeq; - while (refSeq.getDatasetSequence() != null) - { - refSeq = refSeq.getDatasetSequence(); - } - length = _refSeq.getEnd() - _refSeq.getStart() + 1; - } + @SuppressWarnings("unchecked") public PAEContactMatrix(SequenceI _refSeq, Map pae_obj) + throws FileFormatException { setRefSeq(_refSeq); // convert the lists to primitive arrays and store - + if (!MapUtils.containsAKey(pae_obj, "predicted_aligned_error", "pae")) { parse_version_1_pAE(pae_obj); @@ -59,32 +65,51 @@ public class PAEContactMatrix implements ContactMatrixI parse_version_2_pAE(pae_obj); } } + /** * construct a sequence associated PAE matrix directly from a float array + * * @param _refSeq * @param matrix */ public PAEContactMatrix(SequenceI _refSeq, float[][] matrix) { setRefSeq(_refSeq); - maxcol=0; - for (float[] row:matrix) + maxcol = 0; + for (float[] row : matrix) { - if (row.length>maxcol) + if (row.length > maxcol) { - maxcol=row.length; + maxcol = row.length; } - maxscore=row[0]; - for (float f:row) + maxscore = row[0]; + for (float f : row) { - if (maxscore pae_obj) { - // this is never going to be reached by the integer rounding.. or is it ? - maxscore = ((Double) MapUtils.getFirst(pae_obj, - "max_predicted_aligned_error", "max_pae")).floatValue(); + maxscore = -1; + // look for a maxscore element - if there is one... + try + { + // this is never going to be reached by the integer rounding.. or is it ? + maxscore = ((Double) MapUtils.getFirst(pae_obj, + "max_predicted_aligned_error", "max_pae")).floatValue(); + } catch (Throwable t) + { + // ignore if a key is not found. + } List> scoreRows = ((List>) MapUtils - .getFirst(pae_obj, "predicted_aligned_error", "pae")) - ; + .getFirst(pae_obj, "predicted_aligned_error", "pae")); elements = new float[scoreRows.size()][scoreRows.size()]; int row = 0, col = 0; - for (List scoreRow:scoreRows) + for (List scoreRow : scoreRows) { Iterator scores = scoreRow.iterator(); while (scores.hasNext()) { Object d = scores.next(); if (d instanceof Double) + { elements[row][col++] = ((Double) d).longValue(); + } else - elements[row][col++] = (float) ((Long)d).longValue(); + { + elements[row][col++] = (float) ((Long) d).longValue(); + } + + if (maxscore < elements[row][col - 1]) + { + maxscore = elements[row][col - 1]; + } } row++; col = 0; @@ -134,10 +175,26 @@ public class PAEContactMatrix implements ContactMatrixI // dataset refSeq Iterator rows = ((List) pae_obj.get("residue1")).iterator(); Iterator cols = ((List) pae_obj.get("residue2")).iterator(); + // two pass - to allocate the elements array + while (rows.hasNext()) + { + int row = rows.next().intValue(); + int col = cols.next().intValue(); + if (maxrow < row) + { + maxrow = row; + } + if (maxcol < col) + { + maxcol = col; + } + + } + rows = ((List) pae_obj.get("residue1")).iterator(); + cols = ((List) pae_obj.get("residue2")).iterator(); Iterator scores = ((List) pae_obj.get("distance")) .iterator(); - // assume square matrix - elements = new float[length][length]; + elements = new float[maxrow][maxcol]; while (scores.hasNext()) { float escore = scores.next().floatValue(); @@ -159,9 +216,9 @@ public class PAEContactMatrix implements ContactMatrixI } @Override - public ContactListI getContactList(final int _column) + public ContactListI getContactList(final int column) { - if (_column < 0 || _column >= elements.length) + if (column < 0 || column >= elements.length) { return null; } @@ -171,133 +228,117 @@ public class PAEContactMatrix implements ContactMatrixI @Override public int getPosition() { - return _column; + return column; } @Override public int getContactHeight() { - return maxcol-1; + return maxcol - 1; } @Override - public double getContactAt(int column) + public double getContactAt(int mcolumn) { - if (column < 0 || column >= elements[_column].length) + if (mcolumn < 0 || mcolumn >= elements[column].length) { return -1; } - return elements[_column][column]; + return elements[column][mcolumn]; } }); } @Override - public float getMin() + protected double getElementAt(int _column, int i) { - return 0; + return elements[_column][i]; } @Override - public float getMax() - { - return maxscore; - } - - @Override - public boolean hasReferenceSeq() + public float getMin() { - return (refSeq != null); + return 0; } @Override - public SequenceI getReferenceSeq() + public float getMax() { - return refSeq; + return maxscore; } @Override public String getAnnotDescr() { - return "Predicted Alignment Error for " + refSeq.getName(); + return "Predicted Alignment Error" + + ((refSeq == null) ? "" : (" for " + refSeq.getName())); } @Override public String getAnnotLabel() { - return "pAE Matrix"; + StringBuilder label = new StringBuilder("PAE Matrix"); + // if (this.getReferenceSeq() != null) + // { + // label.append(":").append(this.getReferenceSeq().getDisplayId(false)); + // } + return label.toString(); } - public static final String PAEMATRIX="PAE_MATRIX"; + public static final String PAEMATRIX = "PAE_MATRIX"; + @Override public String getType() { return PAEMATRIX; } + @Override public int getWidth() { return length; } + @Override public int getHeight() { return length; } - List groups=null; - @Override - public boolean hasGroups() - { - return groups!=null; - } - String newick=null; - public String getNewickString() - { - return newick; - } - public void makeGroups(float thresh,boolean abs) + public static void validateContactMatrixFile(String fileName) + throws FileFormatException, IOException { - AverageDistanceEngine clusterer = new AverageDistanceEngine(null, null, this); - double height = clusterer.findHeight(clusterer.getTopNode()); - newick = new jalview.io.NewickFile(clusterer.getTopNode(),false,true).print(); - - Console.trace("Newick string\n"+newick); - - List nodegroups; - if (abs ? height > thresh : 0 < thresh && thresh < 1) + FileInputStream infile = null; + try { - float cut = abs ? (float) (thresh / height) : thresh; - Console.debug("Threshold "+cut+" for height="+height); - - nodegroups = clusterer.groupNodes(cut); + infile = new FileInputStream(new File(fileName)); + } catch (Throwable t) + { + new IOException("Couldn't open " + fileName, t); } - else + JSONObject paeDict = null; + try { - nodegroups = new ArrayList(); - nodegroups.add(clusterer.getTopNode()); + paeDict = EBIAlfaFold.parseJSONtoPAEContactMatrix(infile); + } catch (Throwable t) + { + new FileFormatException("Couldn't parse " + fileName + + " as a JSON dict or array containing a dict"); } - groups = new ArrayList<>(); - for (BinaryNode root:nodegroups) + PAEContactMatrix matrix = new PAEContactMatrix( + new SequenceDummy("Predicted"), (Map) paeDict); + if (matrix.getWidth() <= 0) { - BitSet gpset=new BitSet(); - for (BinaryNode leaf:clusterer.findLeaves(root)) - { - gpset.set((Integer)leaf.element()); - } - groups.add(gpset); + throw new FileFormatException( + "No data in PAE matrix read from '" + fileName + "'"); } } - @Override - public BitSet getGroupsFor(int column) + protected PAEContactMatrix newMappableContactMatrix(SequenceI newRefSeq, + MapList newFromMapList) { - for (BitSet gp:groups) { - if (gp.get(column)) - { - return gp; - } - } - return ContactMatrixI.super.getGroupsFor(column); + PAEContactMatrix pae = new PAEContactMatrix(newRefSeq, newFromMapList, + elements, new GroupSet(grps)); + return pae; } }