/**
* Filter a UniProt XML dump file, write entries for organisms of interest to a new
* BufferedWriter. Currently ignores entries for more than one organism.
*
* @param in the UniProt XML dump
* @param out to write output to
* @throws IOException if problem with input or output
*/
public void filter(BufferedReader in, BufferedWriter out) throws IOException {
StringBuffer sb = new StringBuffer();
String line = null;
boolean keep = true;
boolean inOrganism = false;
boolean foundTaxon = false;
while ((line = in.readLine()) != null) {
// quicker to trim whole file first?
String trimmed = StringUtil.trimLeft(line);
if (trimmed.startsWith("<entry")) {
// make sure opening element is included
if (keep) {
out.write(sb.toString());
}
sb = new StringBuffer();
keep = true;
inOrganism = false;
foundTaxon = false;
} else if (trimmed.startsWith("<organism")) {
inOrganism = true;
foundTaxon = false;
} else if (inOrganism && trimmed.startsWith("<dbReference type=\"NCBI Taxonomy")) {
// ignores the possibility of a protein being linked to multiple organisms
foundTaxon = true;
int start = trimmed.indexOf("id=\"") + 4;
String taxonId = trimmed.substring(start, trimmed.indexOf('"', start));
if (!(organisms.contains(taxonId))) {
keep = false;
}
} else if (inOrganism && trimmed.startsWith("</organism") && !foundTaxon) {
// if the organism has no taxon defined then we don't want it
keep = false;
}
if (keep) {
sb.append(StringUtil.escapeBackslash(line) + System.getProperty("line.separator"));
}
}
// catch final entry
if (keep) {
out.write(sb.toString());
} else {
out.write("</uniprot>");
}
out.flush();
}
protected void createFromFile(File f) throws IOException {
// data is in format:
// ZDBID ID1,ID2,ID3
Iterator<?> lineIter =
FormattedTextParser.parseTabDelimitedReader(new BufferedReader(new FileReader(f)));
while (lineIter.hasNext()) {
String[] line = (String[]) lineIter.next();
if (line.length < 2 || line[0].startsWith("#") || !line[0].startsWith(GENE_PATTERN)) {
continue;
}
String zfinId = line[0];
String[] synonyms = StringUtil.split(line[1].trim(), ",");
resolver.addMainIds(taxonId, zfinId, Collections.singleton(zfinId));
resolver.addSynonyms(taxonId, zfinId, new HashSet<String>(Arrays.asList(synonyms)));
}
}
private String newPublication(String codes) throws ObjectStoreException {
String pubRefId = null;
String[] array = codes.split("[|]");
for (int i = 0; i < array.length; i++) {
if (array[i].startsWith("PMID:")) {
String pubMedId = array[i].substring(5);
if (StringUtil.allDigits(pubMedId)) {
pubRefId = publications.get(pubMedId);
if (pubRefId == null) {
Item item = createItem("Publication");
item.setAttribute("pubMedId", pubMedId);
pubRefId = item.getIdentifier();
publications.put(pubMedId, pubRefId);
store(item);
}
return pubRefId;
}
}
}
return null;
}
private void combine(ActionForm form, HttpServletRequest request, String opText) {
HttpSession session = request.getSession();
final InterMineAPI im = SessionMethods.getInterMineAPI(session);
Profile profile = SessionMethods.getProfile(session);
ModifyBagForm mbf = (ModifyBagForm) form;
BagManager bagManager = im.getBagManager();
Map<String, InterMineBag> allBags = bagManager.getBags(profile);
String[] selectedBagNames = mbf.getSelectedBags();
Collection<InterMineBag> selectedBags = getSelectedBags(allBags, selectedBagNames);
String newBagName = NameUtil.validateName(allBags.keySet(), mbf.getNewBagName());
int newBagSize = 0;
try {
if (opText.equals(BagOperations.UNION)) {
newBagSize = BagOperations.union(selectedBags, newBagName, profile, im.getClassKeys());
} else if (opText.equals(BagOperations.INTERSECT)) {
newBagSize = BagOperations.intersect(selectedBags, newBagName, profile, im.getClassKeys());
} else if (opText.equals(BagOperations.SUBTRACT)) {
newBagSize = BagOperations.subtract(selectedBags, newBagName, profile, im.getClassKeys());
}
} catch (IncompatibleTypesException e) {
SessionMethods.recordError(
"You can only perform operations on lists of the same type."
+ " Lists "
+ StringUtil.prettyList(Arrays.asList(selectedBagNames))
+ " do not match.",
session);
return;
} catch (ObjectStoreException e) {
LOG.error(e);
ActionMessage actionMessage = new ActionMessage("An error occurred while saving the list");
recordError(actionMessage, request);
return;
}
if (newBagSize > 0) {
SessionMethods.recordMessage(
"Created list \""
+ newBagName
+ "\" as "
+ opText
+ " of "
+ StringUtil.prettyList(Arrays.asList(selectedBagNames))
+ ".",
session);
// track the list creation
im.getTrackerDelegate()
.trackListCreation(
BagOperations.getCommonBagType(selectedBags),
newBagSize,
ListBuildMode.OPERATION,
profile,
session.getId());
} else {
SessionMethods.recordError(
opText
+ " operation on lists "
+ StringUtil.prettyList(Arrays.asList(selectedBagNames))
+ " produced no results.",
session);
}
}
/**
* Configures a datasource from a Properties object
*
* @param props the properties for configuring the Database
* @throws ClassNotFoundException if the class given in the properties file cannot be found
* @throws IllegalArgumentException if the configuration properties are empty
* @throws NullPointerException if props is null
*/
protected void configure(Properties props) throws ClassNotFoundException {
if (props == null) {
throw new NullPointerException("Props cannot be null");
}
if (props.size() == 0) {
throw new IllegalArgumentException("No configuration details");
}
Properties subProps = new Properties();
for (Map.Entry<Object, Object> entry : props.entrySet()) {
String propertyName = (String) entry.getKey();
String propertyValue = (String) entry.getValue();
Field field = null;
// Get the first part of the string - this is the attribute we are taking about
String attribute = propertyName;
String subAttribute = "";
int index = propertyName.indexOf(".");
if (index != -1) {
attribute = propertyName.substring(0, index);
subAttribute = propertyName.substring(index + 1);
}
try {
field = Database.class.getDeclaredField(attribute);
} catch (Exception e) {
LOG.warn("Ignoring field for Database: " + attribute);
// Ignore this property - no such field
continue;
}
if ("class".equals(subAttribute)) {
// make a new instance of this class for this attribute
Class<?> clazz = Class.forName(propertyValue.toString());
Object obj;
try {
obj = clazz.newInstance();
} catch (Exception e) {
throw new ClassNotFoundException(
"Cannot instantiate class " + clazz.getName() + " " + e.getMessage());
}
// Set the field to this newly instantiated class
try {
field.set(this, obj);
} catch (Exception e) {
continue;
}
} else if ("".equals(subAttribute)) {
// Set this attribute directly
try {
field.set(this, propertyValue);
} catch (Exception e) {
continue;
}
} else {
// Set parameters on the attribute
Method m = null;
// Set this configuration parameter on the DataSource;
try {
// Strings first
Object o = field.get(this);
// Sometimes the class will not have been instantiated yet
if (o == null) {
subProps.put(propertyName, propertyValue);
continue;
}
Class<?> clazz = o.getClass();
m =
clazz.getMethod(
"set" + StringUtil.capitalise(subAttribute), new Class[] {String.class});
if (m != null) {
m.invoke(field.get(this), new Object[] {propertyValue});
}
// now integers
} catch (Exception e) {
// Don't do anything - either the method not found or cannot be invoked
}
try {
if (m == null) {
m =
field
.get(this)
.getClass()
.getMethod(
"set" + StringUtil.capitalise(subAttribute), new Class[] {int.class});
if (m != null) {
m.invoke(field.get(this), new Object[] {Integer.valueOf(propertyValue.toString())});
}
}
} catch (Exception e) {
// Don't do anything - either the method not found or cannot be invoked
}
}
if (subProps.size() > 0) {
configure(subProps);
}
}
}
/**
* Write code for getters and setters for given field.
*
* @param field descriptor for field
* @param fieldPresent true if this class has the associated field
* @return string with generated java code
*/
protected String generateGetSet(FieldDescriptor field, boolean fieldPresent) {
String name = field.getName();
String type = getType(field);
StringBuffer sb = new StringBuffer();
// Get method
sb.append(INDENT)
.append("public ")
.append(type)
.append(" get")
.append(StringUtil.reverseCapitalisation(name))
.append("()");
if (!fieldPresent) {
sb.append(";" + ENDL);
} else {
sb.append(" { ");
if ((field instanceof ReferenceDescriptor) && (!(field instanceof CollectionDescriptor))) {
// This is an object reference.
sb.append("if (")
.append(name)
.append(" instanceof org.intermine.objectstore.proxy.ProxyReference) { return ")
.append("((")
.append(type)
.append(") ((org.intermine.objectstore.proxy.ProxyReference) ")
.append(name)
.append(").getObject()); }; return (")
.append(type)
.append(") ")
.append(name)
.append("; }" + ENDL);
} else {
sb.append("return ").append(name).append("; }" + ENDL);
}
}
// Set method
sb.append(INDENT)
.append("public void ")
.append("set")
.append(StringUtil.reverseCapitalisation(name))
.append("(final ")
.append(type)
.append(" ")
.append(name)
.append(")");
if (!fieldPresent) {
sb.append(";" + ENDL);
} else {
sb.append(" { ").append("this.").append(name).append(" = ").append(name).append("; }" + ENDL);
}
if (field instanceof ReferenceDescriptor) {
if (field instanceof CollectionDescriptor) {
sb.append(INDENT)
.append("public void add")
.append(StringUtil.reverseCapitalisation(name))
.append("(final ")
.append(((CollectionDescriptor) field).getReferencedClassDescriptor().getName())
.append(" arg)");
if (fieldPresent) {
sb.append(" { ").append(name).append(".add(arg); }" + ENDL);
} else {
sb.append(";" + ENDL);
}
} else {
// This is an object reference.
sb.append(INDENT)
.append("public void proxy")
.append(StringUtil.reverseCapitalisation(name))
.append("(final org.intermine.objectstore.proxy.ProxyReference ")
.append(name)
.append(")");
if (fieldPresent) {
sb.append(" { this.").append(name).append(" = ").append(name).append("; }" + ENDL);
} else {
sb.append(";" + ENDL);
}
sb.append(INDENT)
.append("public org.intermine.model.InterMineObject proxGet")
.append(StringUtil.reverseCapitalisation(name))
.append("()");
if (fieldPresent) {
sb.append(" { return ").append(name).append("; }" + ENDL);
} else {
sb.append(";" + ENDL);
}
}
}
return sb.toString();
}
/** {@inheritDoc} */
public void process(Reader reader) throws Exception {
// Create a chromosome
Item chromosome = createItem("Chromosome");
chromosome.setAttribute("primaryIdentifier", CHROMOSOME_PID);
store(chromosome);
@SuppressWarnings("rawtypes")
Iterator lineIter = FormattedTextParser.parseTabDelimitedReader(reader);
while (lineIter.hasNext()) {
String[] line = (String[]) lineIter.next();
// remove header line
if (!line[0].equals(HEADER_LINE)) {
String ecogeneId = line[0];
String geneName = line[1];
String eCK = line[2];
String swissProtId = line[3];
String wisconsinGenBankId = line[4];
String genBankProteinId = line[5];
String genoBaseId = line[6];
String type = line[7];
String strand = line[8];
String start = line[9];
String end = line[10];
String synonym = line[11];
Set<String> symSet = new TreeSet<String>();
if (!eCK.equals(NULL_STRING)) {
symSet.add(eCK);
}
if (!genoBaseId.equals(NULL_STRING)) {
symSet.addAll(Arrays.asList(StringUtil.split(genoBaseId, "; ")));
}
if (!synonym.equals(NONE_STRING)) {
symSet.addAll(Arrays.asList(synonym.split(", ")));
}
if (type.equals(TYPE_GENE)) {
Item gene = createItem("Gene");
gene.setReference("chromosome", chromosome);
gene.setReference("organism", getOrganism(ECOLI_TAXON));
gene.setAttribute("primaryIdentifier", ecogeneId);
gene.setAttribute("secondaryIdentifier", wisconsinGenBankId);
gene.setAttribute("name", geneName);
gene.setAttribute("symbol", geneName);
if (symSet.size() > 0) {
for (String sym : symSet) {
createSynonym(gene, sym, true);
}
}
if (!swissProtId.equals(NULL_STRING)) {
if (proteinMap.containsKey(swissProtId)) {
// Reference a protein to a gene (a gene has proteins
// collection)
gene.addToCollection("proteins", proteinMap.get(swissProtId));
} else {
Item protein = createItem("Protein");
protein.setAttribute("primaryAccession", swissProtId);
// NCBI Protein id, remove "g"
protein.setAttribute("secondaryIdentifier", genBankProteinId.substring(1));
gene.addToCollection("proteins", protein);
store(protein);
proteinMap.put(swissProtId, protein);
}
}
// Create chromosome location
if (start.matches(DIGIT_REGEX) && end.matches(DIGIT_REGEX)) {
Item location = createItem("Location");
location.setAttribute("start", start);
location.setAttribute("end", end);
location.setReference("feature", gene);
location.setReference("locatedOn", chromosome);
if (strand.equals(CLOCKWISE)) {
location.setAttribute("strand", "+1");
} else if (strand.equals(COUNTER_CLOCKWISE)) {
location.setAttribute("strand", "-1");
} else {
location.setAttribute("strand", "0");
}
gene.setReference("chromosomeLocation", location);
store(location);
}
store(gene);
} else if (type.equals(TYPE_RNA)) { // TODO code refactory
Item rna = createItem("NcRNA");
rna.setReference("chromosome", chromosome);
rna.setReference("organism", getOrganism(ECOLI_TAXON));
rna.setAttribute("primaryIdentifier", ecogeneId);
rna.setAttribute("secondaryIdentifier", wisconsinGenBankId);
rna.setAttribute("name", geneName);
rna.setAttribute("symbol", geneName);
if (symSet.size() > 0) {
for (String sym : symSet) {
createSynonym(rna, sym, true);
}
}
// Create chromosome location
if (start.matches(DIGIT_REGEX) && end.matches(DIGIT_REGEX)) {
Item location = createItem("Location");
location.setAttribute("start", start);
location.setAttribute("end", end);
location.setReference("feature", rna);
location.setReference("locatedOn", chromosome);
if (strand.equals(CLOCKWISE)) {
location.setAttribute("strand", "+1");
} else if (strand.equals(COUNTER_CLOCKWISE)) {
location.setAttribute("strand", "-1");
} else {
location.setAttribute("strand", "0");
}
rna.setReference("chromosomeLocation", location);
store(location);
}
store(rna);
}
}
}
}
/**
* Obtain the pubmed esummary information for the publications
*
* @param ids the pubMedIds of the publications
* @return a Reader for the information
* @throws Exception if an error occurs
*/
protected Reader getReader(Set<Integer> ids) throws Exception {
String urlString = EFETCH_URL + StringUtil.join(ids, ",");
System.err.println("retrieving: " + urlString);
return new BufferedReader(new InputStreamReader(new URL(urlString).openStream()));
}
/**
* Return a Set of PrimaryKeys relevant to a given Source for a ClassDescriptor. The Set contains
* all the primary keys that exist on a particular class that are used by the source, without
* performing any recursion. The Model.getClassDescriptorsForClass() method is recommended if you
* wish for all the primary keys of the class' parents as well.
*
* @param cld the ClassDescriptor
* @param source the Source
* @param os the ObjectStore that these PrimaryKeys are used in, for creating indexes
* @return a Set of PrimaryKeys
*/
public static Set<PrimaryKey> getPrimaryKeys(ClassDescriptor cld, Source source, ObjectStore os) {
GetPrimaryKeyCacheKey key = new GetPrimaryKeyCacheKey(cld, source);
synchronized (getPrimaryKeyCache) {
Set<PrimaryKey> keySet = getPrimaryKeyCache.get(key);
if (keySet == null) {
keySet = new LinkedHashSet<PrimaryKey>();
Properties keys = getKeyProperties(source);
if (keys != null) {
if (!verifiedSources.contains(source)) {
String packageNameWithDot =
cld.getName().substring(0, cld.getName().lastIndexOf('.') + 1);
LOG.info(
"Verifying primary key config for source "
+ source
+ ", packageName = "
+ packageNameWithDot);
for (Map.Entry<Object, Object> entry : keys.entrySet()) {
String cldName = (String) entry.getKey();
String keyList = (String) entry.getValue();
if (!cldName.contains(".")) {
ClassDescriptor iCld =
cld.getModel().getClassDescriptorByName(packageNameWithDot + cldName);
if (iCld != null) {
Map<String, PrimaryKey> map = PrimaryKeyUtil.getPrimaryKeys(iCld);
String[] tokens = keyList.split(",");
for (int i = 0; i < tokens.length; i++) {
String token = tokens[i].trim();
if (map.get(token) == null) {
throw new IllegalArgumentException(
"Primary key "
+ token
+ " for class "
+ cldName
+ " required by datasource "
+ source.getName()
+ " in "
+ source.getName()
+ "_keys.properties is not defined in "
+ cld.getModel().getName()
+ "_keyDefs.properties");
}
}
} else {
LOG.warn(
"Ignoring entry for "
+ cldName
+ " in file "
+ cld.getModel().getName()
+ "_keyDefs.properties - not in model!");
}
}
}
verifiedSources.add(source);
}
Map<String, PrimaryKey> map = PrimaryKeyUtil.getPrimaryKeys(cld);
String cldName = TypeUtil.unqualifiedName(cld.getName());
String keyList = (String) keys.get(cldName);
if (keyList != null) {
String[] tokens = keyList.split(",");
for (int i = 0; i < tokens.length; i++) {
String token = tokens[i].trim();
if (map.get(token) == null) {
throw new IllegalArgumentException(
"Primary key "
+ token
+ " for class "
+ cld.getName()
+ " required by data source "
+ source.getName()
+ " in "
+ source.getName()
+ "_keys.properties is not defined in "
+ cld.getModel().getName()
+ "_keyDefs.properties");
} else {
keySet.add(map.get(token));
}
}
}
for (Map.Entry<Object, Object> entry : keys.entrySet()) {
String propKey = (String) entry.getKey();
String fieldList = (String) entry.getValue();
int posOfDot = propKey.indexOf('.');
if (posOfDot > 0) {
String propCldName = propKey.substring(0, posOfDot);
if (cldName.equals(propCldName)) {
String keyName = propKey.substring(posOfDot + 1);
PrimaryKey pk = new PrimaryKey(keyName, fieldList, cld);
if (!keySet.contains(pk)) {
keySet.add(pk);
if (os instanceof ObjectStoreInterMineImpl) {
ObjectStoreInterMineImpl osimi = (ObjectStoreInterMineImpl) os;
DatabaseSchema schema = osimi.getSchema();
ClassDescriptor tableMaster = schema.getTableMaster(cld);
String tableName = DatabaseUtil.getTableName(tableMaster);
List<String> fields = new ArrayList<String>();
for (String field : pk.getFieldNames()) {
String colName = DatabaseUtil.generateSqlCompatibleName(field);
if (tableMaster.getReferenceDescriptorByName(field, true) != null) {
colName += "id";
}
fields.add(colName);
}
String sql =
"CREATE INDEX "
+ tableName
+ "__"
+ keyName
+ " ON "
+ tableName
+ " ("
+ StringUtil.join(fields, ", ")
+ ")";
System.out.println("Creating index: " + sql);
LOG.info("Creating index: " + sql);
Connection conn = null;
try {
conn = osimi.getConnection();
conn.createStatement().execute(sql);
} catch (SQLException e) {
LOG.warn("Index creation failed", e);
} finally {
if (conn != null) {
osimi.releaseConnection(conn);
}
}
}
}
}
}
}
} else {
throw new IllegalArgumentException(
"Unable to find keys for source "
+ source.getName()
+ " in file "
+ source.getName()
+ "_keys.properties");
}
getPrimaryKeyCache.put(key, keySet);
}
return keySet;
}
}
