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 JSONObject paeDict = parseJSONtoPAEContactMatrix(pae_input);
388 Console.debug("JSON file did not parse properly.");
391 ContactMatrixI matrix = new PAEContactMatrix(sequence,
392 (Map<String, Object>) paeDict);
394 AlignmentAnnotation cmannot = sequence.addContactList(matrix);
395 pdbAlignment.addAnnotation(cmannot);
400 public static JSONObject parseJSONtoPAEContactMatrix(
401 InputStream pae_input) throws IOException, ParseException
403 Object paeJson = Platform.parseJSON(pae_input);
404 JSONObject paeDict = null;
405 if (paeJson instanceof JSONObject)
407 paeDict = (JSONObject) paeJson;
409 else if (paeJson instanceof JSONArray)
411 JSONArray jsonArray = (JSONArray) paeJson;
412 if (jsonArray.size() > 0)
413 paeDict = (JSONObject) jsonArray.get(0);
419 public static boolean importPaeJSONAsContactMatrixToStructure(
420 StructureMapping[] smArray, InputStream paeInput)
421 throws IOException, ParseException
423 boolean someDone = false;
424 for (StructureMapping sm : smArray)
426 boolean thisDone = importPaeJSONAsContactMatrixToStructure(sm,
428 someDone |= thisDone;
433 public static boolean importPaeJSONAsContactMatrixToStructure(
434 StructureMapping sm, InputStream paeInput)
435 throws IOException, ParseException
438 JSONObject pae_obj = parseJSONtoPAEContactMatrix(paeInput);
441 Console.debug("JSON file did not parse properly.");
445 ContactMatrixI matrix = new PAEContactMatrix(sm.getSequence(),
446 (Map<String, Object>) pae_obj);
448 AlignmentAnnotation cmannot = sm.getSequence().addContactList(matrix);
449 // sm.getSequence().addAlignmentAnnotation(cmannot);
450 sm.transfer(cmannot);
453 StructureSelectionManager ssm = StructureSelectionManager
454 .getStructureSelectionManager(Desktop.instance);
455 List<AlignedCodonFrame> acfList = ssm.getSequenceMappings();
461 * general purpose structure importer - designed to yield alignment useful for
462 * transfer of annotation to associated sequences
464 * @param alphaFoldCif
473 public static AlignmentI importDownloadedStructureFromUrl(
474 String alphaFoldCif, File tmpFile, String id, String chain,
475 String dbSource, String dbVersion) throws Exception
477 String file = tmpFile.getAbsolutePath();
478 // todo get rid of Type and use FileFormatI instead?
479 FileFormatI fileFormat = FileFormat.MMCif;
480 AlignmentI pdbAlignment = new FormatAdapter().readFile(tmpFile,
481 DataSourceType.FILE, fileFormat);
482 if (pdbAlignment != null)
484 List<SequenceI> toremove = new ArrayList<SequenceI>();
485 for (SequenceI pdbcs : pdbAlignment.getSequences())
489 for (PDBEntry pid : pdbcs.getAllPDBEntries())
491 if (pid.getFile() == file)
493 chid = pid.getChainCode();
497 if (chain == null || (chid != null && (chid.equals(chain)
498 || chid.trim().equals(chain.trim())
499 || (chain.trim().length() == 0 && chid.equals("_")))))
501 // FIXME seems to result in 'PDB|1QIP|1qip|A' - 1QIP is redundant.
502 // TODO: suggest simplify naming to 1qip|A as default name defined
503 pdbcs.setName(id + SEPARATOR + pdbcs.getName());
504 // Might need to add more metadata to the PDBEntry object
507 * PDBEntry entry = new PDBEntry(); // Construct the PDBEntry
508 * entry.setId(id); if (entry.getProperty() == null)
509 * entry.setProperty(new Hashtable());
510 * entry.getProperty().put("chains", pdbchain.id + "=" +
511 * sq.getStart() + "-" + sq.getEnd());
512 * sq.getDatasetSequence().addPDBId(entry);
515 // We make a DBRefEtntry because we have obtained the PDB file from
518 // JBPNote - PDB DBRefEntry should also carry the chain and mapping
520 if (dbSource != null)
522 DBRefEntry dbentry = new DBRefEntry(dbSource,
524 dbVersion, (chid == null ? id : id + chid));
526 pdbcs.addDBRef(dbentry);
527 // update any feature groups
528 List<SequenceFeature> allsf = pdbcs.getFeatures()
530 List<SequenceFeature> newsf = new ArrayList<SequenceFeature>();
531 if (allsf != null && allsf.size() > 0)
533 for (SequenceFeature f : allsf)
535 if (file.equals(f.getFeatureGroup()))
537 f = new SequenceFeature(f, f.type, f.begin, f.end, id,
542 pdbcs.setSequenceFeatures(newsf);
548 // mark this sequence to be removed from the alignment
549 // - since it's not from the right chain
553 // now remove marked sequences
554 for (SequenceI pdbcs : toremove)
556 pdbAlignment.deleteSequence(pdbcs);
557 if (pdbcs.getAnnotation() != null)
559 for (AlignmentAnnotation aa : pdbcs.getAnnotation())
561 pdbAlignment.deleteAnnotation(aa);
572 * @see jalview.ws.DbSourceProxy#isValidReference(java.lang.String)
575 public boolean isValidReference(String accession)
577 Regex r = getAccessionValidator();
578 return r.search(accession.trim());
585 public String getTestQuery()
587 return "AF-O15552-F1";
591 public String getDbName()
593 return "ALPHAFOLD"; // getDbSource();
603 * Returns a descriptor for suitable feature display settings with
605 * <li>ResNums or insertions features visible</li>
606 * <li>insertions features coloured red</li>
607 * <li>ResNum features coloured by label</li>
608 * <li>Insertions displayed above (on top of) ResNums</li>
612 public FeatureSettingsModelI getFeatureColourScheme()
614 return new PDBFeatureSettings();