3 * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
4 * Copyright (C) $$Year-Rel$$ The Jalview Authors
6 * This file is part of Jalview.
8 * Jalview is free software: you can redistribute it and/or
9 * modify it under the terms of the GNU General Public License
10 * as published by the Free Software Foundation, either version 3
11 * of the License, or (at your option) any later version.
13 * Jalview is distributed in the hope that it will be useful, but
14 * WITHOUT ANY WARRANTY; without even the implied warranty
15 * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
16 * PURPOSE. See the GNU General Public License for more details.
18 * You should have received a copy of the GNU General Public License
19 * along with Jalview. If not, see <http://www.gnu.org/licenses/>.
20 * The Jalview Authors are detailed in the 'AUTHORS' file.
22 package jalview.ws.dbsources;
25 import java.io.FileInputStream;
26 import java.io.FileNotFoundException;
27 import java.io.IOException;
28 import java.io.InputStream;
29 import java.util.ArrayList;
30 import java.util.List;
33 import org.json.simple.JSONArray;
34 import org.json.simple.JSONObject;
35 import org.json.simple.parser.ParseException;
37 import com.stevesoft.pat.Regex;
39 import jalview.api.FeatureSettingsModelI;
40 import jalview.bin.Console;
41 import jalview.datamodel.AlignedCodonFrame;
42 import jalview.datamodel.AlignmentAnnotation;
43 import jalview.datamodel.AlignmentI;
44 import jalview.datamodel.ContactMatrixI;
45 import jalview.datamodel.DBRefEntry;
46 import jalview.datamodel.PDBEntry;
47 import jalview.datamodel.SequenceFeature;
48 import jalview.datamodel.SequenceI;
49 import jalview.gui.Desktop;
50 import jalview.io.DataSourceType;
51 import jalview.io.FileFormat;
52 import jalview.io.FileFormatI;
53 import jalview.io.FormatAdapter;
54 import jalview.io.PDBFeatureSettings;
55 import jalview.structure.StructureMapping;
56 import jalview.structure.StructureSelectionManager;
57 import jalview.util.MessageManager;
58 import jalview.util.Platform;
59 import jalview.ws.datamodel.alphafold.PAEContactMatrix;
60 import jalview.ws.utils.UrlDownloadClient;
66 public class EBIAlfaFold extends EbiFileRetrievedProxy
68 private static final String SEPARATOR = "|";
70 private static final String COLON = ":";
72 private static final int PDB_ID_LENGTH = 4;
74 private static String AF_VERSION = "3";
84 * @see jalview.ws.DbSourceProxy#getAccessionSeparator()
87 public String getAccessionSeparator()
95 * @see jalview.ws.DbSourceProxy#getAccessionValidator()
98 public Regex getAccessionValidator()
100 Regex validator = new Regex("(AF-[A-Z]+[0-9]+[A-Z0-9]+-F1)");
101 validator.setIgnoreCase(true);
108 * @see jalview.ws.DbSourceProxy#getDbSource()
111 public String getDbSource()
119 * @see jalview.ws.DbSourceProxy#getDbVersion()
122 public String getDbVersion()
127 public static String getAlphaFoldCifDownloadUrl(String id, String vnum)
129 if (vnum == null || vnum.length() == 0)
133 return "https://alphafold.ebi.ac.uk/files/" + id + "-model_v" + vnum
137 public static String getAlphaFoldPaeDownloadUrl(String id, String vnum)
139 if (vnum == null || vnum.length() == 0)
143 return "https://alphafold.ebi.ac.uk/files/" + id
144 + "-predicted_aligned_error_v" + vnum + ".json";
150 * @see jalview.ws.DbSourceProxy#getSequenceRecords(java.lang.String[])
153 public AlignmentI getSequenceRecords(String queries) throws Exception
155 return getSequenceRecords(queries, null);
158 public AlignmentI getSequenceRecords(String queries, String retrievalUrl)
161 AlignmentI pdbAlignment = null;
164 if (queries.indexOf(COLON) > -1)
166 chain = queries.substring(queries.indexOf(COLON) + 1);
167 id = queries.substring(0, queries.indexOf(COLON));
174 if (!isValidReference(id))
177 "(AFClient) Ignoring invalid alphafold query: '" + id + "'");
181 String alphaFoldCif = getAlphaFoldCifDownloadUrl(id, AF_VERSION);
182 if (retrievalUrl != null)
184 alphaFoldCif = retrievalUrl;
189 File tmpFile = File.createTempFile(id, ".cif");
190 Console.debug("Retrieving structure file for " + id + " from "
192 UrlDownloadClient.download(alphaFoldCif, tmpFile);
194 // may not need this check ?
195 file = tmpFile.getAbsolutePath();
201 pdbAlignment = importDownloadedStructureFromUrl(alphaFoldCif, tmpFile,
202 id, chain, getDbSource(), getDbVersion());
204 if (pdbAlignment == null || pdbAlignment.getHeight() < 1)
206 throw new Exception(MessageManager.formatMessage(
207 "exception.no_pdb_records_for_chain", new String[]
208 { id, ((chain == null) ? "' '" : chain) }));
210 // done during structure retrieval
211 // retrieve_AlphaFold_pAE(id, pdbAlignment, retrievalUrl);
213 } catch (Exception ex) // Problem parsing PDB file
222 * get an alphafold pAE for the given id, and add it to sequence 0 in
223 * pdbAlignment (assuming it came from structurefile parser).
226 * @param pdbAlignment
227 * @param retrievalUrl
228 * - URL of .mmcif from EBI-AlphaFold - will be used to generate the
229 * pAE URL automatically
230 * @throws IOException
233 public static void retrieve_AlphaFold_pAE(String id,
234 AlignmentI pdbAlignment, String retrievalUrl) throws IOException
236 // import PAE as contact matrix - assume this will work if there was a
238 String paeURL = getAlphaFoldPaeDownloadUrl(id, AF_VERSION);
240 if (retrievalUrl != null)
242 // manufacture the PAE url from a url like ...-model-vN.cif
243 paeURL = retrievalUrl.replace("model", "predicted_aligned_error")
244 .replace(".cif", ".json");
250 pae = File.createTempFile(id == null ? "af_pae" : id, "pae_json");
251 } catch (IOException e)
255 Console.debug("Downloading pae from " + paeURL + " to " + pae.toString()
257 UrlDownloadClient.download(paeURL, pae);
258 addAlphaFoldPAEToSequence(pdbAlignment, pae, 0, null);
261 public static void addAlphaFoldPAEToSequence(AlignmentI pdbAlignment,
262 File pae, int index, String seqId)
264 addAlphaFoldPAE(pdbAlignment, pae, index, seqId, false, false);
267 public static void addAlphaFoldPAEToStructure(AlignmentI pdbAlignment,
268 File pae, int index, String structIdOrFile, boolean isStructId)
270 addAlphaFoldPAE(pdbAlignment, pae, index, structIdOrFile, true,
274 public static void addAlphaFoldPAE(AlignmentI pdbAlignment, File pae,
275 int index, String id, boolean isStruct, boolean isStructId)
277 FileInputStream paeInput = null;
280 paeInput = new FileInputStream(pae);
281 } catch (FileNotFoundException e)
284 "Could not find pAE file '" + pae.getAbsolutePath() + "'", e);
290 StructureSelectionManager ssm = StructureSelectionManager
291 .getStructureSelectionManager(Desktop.instance);
294 String structFile = isStructId ? ssm.findFileForPDBId(id) : id;
296 StructureMapping[] smArray = ssm.getMapping(structFile);
300 if (!importPaeJSONAsContactMatrixToStructure(smArray, paeInput))
302 Console.warn("Could not import contact matrix from '"
303 + pae.getAbsolutePath() + "' to structure.");
305 } catch (IOException e1)
307 Console.error("Error when importing pAE file '"
308 + pae.getAbsolutePath() + "'", e1);
309 } catch (ParseException e2)
311 Console.error("Error when parsing pAE file '"
312 + pae.getAbsolutePath() + "'", e2);
319 // attach to sequence?!
322 if (!importPaeJSONAsContactMatrixToSequence(pdbAlignment, paeInput,
325 Console.warn("Could not import contact matrix from '"
326 + pae.getAbsolutePath() + "' to sequence.");
328 } catch (IOException e1)
330 Console.error("Error when importing pAE file '"
331 + pae.getAbsolutePath() + "'", e1);
332 } catch (ParseException e2)
334 Console.error("Error when parsing pAE file '"
335 + pae.getAbsolutePath() + "'", e2);
342 * parses the given pAE matrix and adds it to sequence 0 in the given
345 * @param pdbAlignment
347 * @return true if there was a pAE matrix added
348 * @throws ParseException
349 * @throws IOException
352 public static boolean importPaeJSONAsContactMatrixToSequence(
353 AlignmentI pdbAlignment, InputStream pae_input)
354 throws IOException, ParseException
356 return importPaeJSONAsContactMatrixToSequence(pdbAlignment, pae_input,
360 public static boolean importPaeJSONAsContactMatrixToSequence(
361 AlignmentI pdbAlignment, InputStream pae_input, int index,
362 String seqId) throws IOException, ParseException
364 SequenceI sequence = null;
367 int seqToGet = index > 0 ? index : 0;
368 sequence = pdbAlignment.getSequenceAt(seqToGet);
370 if (sequence == null)
372 SequenceI[] sequences = pdbAlignment.findSequenceMatch(seqId);
373 if (sequences == null || sequences.length < 1)
375 Console.warn("Could not find sequence with id '" + seqId
376 + "' to attach pAE matrix to. Ignoring matrix.");
381 sequence = sequences[0]; // just use the first sequence with this seqId
385 Object paeJson = Platform.parseJSON(pae_input);
386 JSONObject paeDict = null;
387 if (paeJson instanceof JSONObject)
389 paeDict = (JSONObject) paeJson;
391 else if (paeJson instanceof JSONArray)
393 JSONArray jsonArray = (JSONArray) paeJson;
394 if (jsonArray.size() > 0)
395 paeDict = (JSONObject) jsonArray.get(0);
400 Console.debug("JSON file did not parse properly.");
403 ContactMatrixI matrix = new PAEContactMatrix(sequence,
404 (Map<String, Object>) paeDict);
406 AlignmentAnnotation cmannot = sequence.addContactList(matrix);
407 pdbAlignment.addAnnotation(cmannot);
412 public static boolean importPaeJSONAsContactMatrixToStructure(
413 StructureMapping[] smArray, InputStream paeInput)
414 throws IOException, ParseException
416 boolean someDone = false;
417 for (StructureMapping sm : smArray)
419 boolean thisDone = importPaeJSONAsContactMatrixToStructure(sm,
421 someDone |= thisDone;
426 public static boolean importPaeJSONAsContactMatrixToStructure(
427 StructureMapping sm, InputStream paeInput)
428 throws IOException, ParseException
431 List<Object> pae_obj = (List<Object>) Platform.parseJSON(paeInput);
434 Console.debug("JSON file did not parse properly.");
438 ContactMatrixI matrix = new PAEContactMatrix(sm.getSequence(),
439 (Map<String, Object>) pae_obj.get(0));
441 AlignmentAnnotation cmannot = sm.getSequence().addContactList(matrix);
442 // sm.getSequence().addAlignmentAnnotation(cmannot);
443 sm.transfer(cmannot);
446 StructureSelectionManager ssm = StructureSelectionManager
447 .getStructureSelectionManager(Desktop.instance);
448 List<AlignedCodonFrame> acfList = ssm.getSequenceMappings();
454 * general purpose structure importer - designed to yield alignment useful for
455 * transfer of annotation to associated sequences
457 * @param alphaFoldCif
466 public static AlignmentI importDownloadedStructureFromUrl(
467 String alphaFoldCif, File tmpFile, String id, String chain,
468 String dbSource, String dbVersion) throws Exception
470 String file = tmpFile.getAbsolutePath();
471 // todo get rid of Type and use FileFormatI instead?
472 FileFormatI fileFormat = FileFormat.MMCif;
473 AlignmentI pdbAlignment = new FormatAdapter().readFile(tmpFile,
474 DataSourceType.FILE, fileFormat);
475 if (pdbAlignment != null)
477 List<SequenceI> toremove = new ArrayList<SequenceI>();
478 for (SequenceI pdbcs : pdbAlignment.getSequences())
482 for (PDBEntry pid : pdbcs.getAllPDBEntries())
484 if (pid.getFile() == file)
486 chid = pid.getChainCode();
490 if (chain == null || (chid != null && (chid.equals(chain)
491 || chid.trim().equals(chain.trim())
492 || (chain.trim().length() == 0 && chid.equals("_")))))
494 // FIXME seems to result in 'PDB|1QIP|1qip|A' - 1QIP is redundant.
495 // TODO: suggest simplify naming to 1qip|A as default name defined
496 pdbcs.setName(id + SEPARATOR + pdbcs.getName());
497 // Might need to add more metadata to the PDBEntry object
500 * PDBEntry entry = new PDBEntry(); // Construct the PDBEntry
501 * entry.setId(id); if (entry.getProperty() == null)
502 * entry.setProperty(new Hashtable());
503 * entry.getProperty().put("chains", pdbchain.id + "=" +
504 * sq.getStart() + "-" + sq.getEnd());
505 * sq.getDatasetSequence().addPDBId(entry);
508 // We make a DBRefEtntry because we have obtained the PDB file from
511 // JBPNote - PDB DBRefEntry should also carry the chain and mapping
513 if (dbSource != null)
515 DBRefEntry dbentry = new DBRefEntry(dbSource,
517 dbVersion, (chid == null ? id : id + chid));
519 pdbcs.addDBRef(dbentry);
520 // update any feature groups
521 List<SequenceFeature> allsf = pdbcs.getFeatures()
523 List<SequenceFeature> newsf = new ArrayList<SequenceFeature>();
524 if (allsf != null && allsf.size() > 0)
526 for (SequenceFeature f : allsf)
528 if (file.equals(f.getFeatureGroup()))
530 f = new SequenceFeature(f, f.type, f.begin, f.end, id,
535 pdbcs.setSequenceFeatures(newsf);
541 // mark this sequence to be removed from the alignment
542 // - since it's not from the right chain
546 // now remove marked sequences
547 for (SequenceI pdbcs : toremove)
549 pdbAlignment.deleteSequence(pdbcs);
550 if (pdbcs.getAnnotation() != null)
552 for (AlignmentAnnotation aa : pdbcs.getAnnotation())
554 pdbAlignment.deleteAnnotation(aa);
565 * @see jalview.ws.DbSourceProxy#isValidReference(java.lang.String)
568 public boolean isValidReference(String accession)
570 Regex r = getAccessionValidator();
571 return r.search(accession.trim());
578 public String getTestQuery()
580 return "AF-O15552-F1";
584 public String getDbName()
586 return "ALPHAFOLD"; // getDbSource();
596 * Returns a descriptor for suitable feature display settings with
598 * <li>ResNums or insertions features visible</li>
599 * <li>insertions features coloured red</li>
600 * <li>ResNum features coloured by label</li>
601 * <li>Insertions displayed above (on top of) ResNums</li>
605 public FeatureSettingsModelI getFeatureColourScheme()
607 return new PDBFeatureSettings();