/**
* Returns the subsample with the given index. In case the subsample doesn't exist, null is
* returned.
*
* @param subsampleId The subsamples index.
* @return
*/
public Subsample getSubsample(Integer subsampleId) {
if (subsamples == null) {
return null;
}
for (Subsample s : subsamples) {
if (s.getSubsampleIndex() == subsampleId) {
return s;
}
}
return null;
}
/**
* Adds the given subsample. In case a subsample with the same index already exists, this subample
* is replaced.
*
* @param s
*/
public void setSubsample(Subsample s) {
if (subsamples == null) {
subsamples = new ArrayList<Subsample>(1);
}
// check if the subsample already exists
for (int i = 0; i < subsamples.size(); i++) {
// if a subsample with the same index already exists, replace it
if (subsamples.get(i).getSubsampleIndex() == s.getSubsampleIndex()) {
subsamples.set(i, s);
return;
}
}
// as the subsample wasn't set, add this one
subsamples.add(s);
}
/**
* Converts the given meta-data to an mzTab formatted string.
*
* @return
*/
public String toMzTab() {
StringBuilder mzTab = new StringBuilder();
if (title != null) {
mzTab.append(createField("title", title));
}
if (description != null) {
mzTab.append(createField("description", description));
}
// sample processing
if (sampleProcessing != null) {
for (Integer i = 1; i <= sampleProcessing.size(); i++) {
mzTab.append(
createField(String.format("sample_processing[%d]", i), sampleProcessing.get(i - 1)));
}
}
// instrument
if (instrument != null) {
for (Integer i = 1; i <= instrument.size(); i++) {
mzTab.append(
createField(String.format("instrument[%d]-name", i), instrument.get(i - 1).getName()));
mzTab.append(
createField(
String.format("instrument[%d]-source", i), instrument.get(i - 1).getSource()));
mzTab.append(
createField(
String.format("instrument[%d]-analyzer", i), instrument.get(i - 1).getAnalyzer()));
mzTab.append(
createField(
String.format("instrument[%d]-detector", i), instrument.get(i - 1).getDetector()));
}
}
// software + software[1-n]-setting
if (software != null) {
for (Integer i = 1; i <= software.size(); i++) {
mzTab.append(createField(String.format("software[%d]", i), software.get(i - 1)));
// write out the settings for the specified software
if (softwareSetting != null && softwareSetting.get(i) != null) {
List<String> settings = softwareSetting.get(i);
for (String s : settings) {
mzTab.append(createField(String.format("software[%d]-setting"), s));
}
}
}
}
// false discovery rate
if (falseDiscoveryRate != null) {
mzTab.append(createField("false_discovery_rate", falseDiscoveryRate));
}
// publication
if (publication != null && publication.size() > 0) {
String string = "";
for (String p : publication) {
string += (string.length() > 1 ? "," : "") + p;
}
mzTab.append(createField("publication", string));
}
// contact
if (contact != null) {
for (int i = 1; i <= contact.size(); i++) {
mzTab.append(
createField(String.format("contact[%d]-name", i), contact.get(i - 1).getName()));
mzTab.append(
createField(
String.format("contact[%d]-affiliation", i), contact.get(i - 1).getAffiliation()));
mzTab.append(
createField(String.format("contact[%d]-email", i), contact.get(i - 1).getEmail()));
}
}
// uri
if (uri != null) {
mzTab.append(createField("uri", uri));
}
// mods
if (mod != null) {
mzTab.append(createField("mod", mod));
}
// quantification method
if (quantificationMethod != null) {
mzTab.append(createField("quantification_method", quantificationMethod));
}
// protein quant unit
if (proteinQuantificationUnit != null) {
mzTab.append(createField("protein-quantification_unit", proteinQuantificationUnit));
}
// peptide quant unit
if (peptideQuantificationUnit != null) {
mzTab.append(createField("peptide-quantification_unit", peptideQuantificationUnit));
}
// ms files
if (msFiles != null) {
List<Integer> ids = new ArrayList<Integer>(msFiles.keySet());
Collections.sort(ids);
for (Integer index : ids) {
mzTab.append(
createField(
String.format("ms_file[%d]-format", index + 1), msFiles.get(index).getFormat()));
mzTab.append(
createField(
String.format("ms_file[%d]-location", index + 1),
msFiles.get(index).getLocation()));
mzTab.append(
createField(
String.format("ms_file[%d]-id_format", index + 1),
msFiles.get(index).getIdFormat()));
}
}
// colunits
if (colunitProtein != null) {
for (String column : colunitProtein.keySet()) {
mzTab.append(
createField(
"colunit-protein",
String.format("%s=%s", column, colunitProtein.get(column).toString())));
}
}
if (colunitPeptide != null) {
for (String column : colunitPeptide.keySet()) {
mzTab.append(
createField(
"colunit-peptide",
String.format("%s=%s", column, colunitPeptide.get(column).toString())));
}
}
if (colunitSmallMolecule != null) {
for (String column : colunitSmallMolecule.keySet()) {
mzTab.append(
createField(
"colunit-small_molecule",
String.format("%s=%s", column, colunitSmallMolecule.get(column).toString())));
}
}
// custom
if (customParams != null) {
for (Param p : customParams) {
mzTab.append(createField("custom", p));
}
}
// species
if (species != null) {
for (int i = 1; i <= species.size(); i++) {
mzTab.append(createField(String.format("species[%d]", i), species.get(i - 1)));
}
}
// tissue
if (tissue != null) {
for (int i = 1; i <= tissue.size(); i++) {
mzTab.append(createField(String.format("tissue[%d]", i), tissue.get(i - 1)));
}
}
// cell_type
if (cellType != null) {
for (int i = 1; i <= cellType.size(); i++) {
mzTab.append(createField(String.format("cell_type[%d]", i), cellType.get(i - 1)));
}
}
// disease
if (disease != null) {
for (int i = 1; i <= disease.size(); i++) {
mzTab.append(createField(String.format("disease[%d]", i), disease.get(i - 1)));
}
}
// subsamples
if (subsamples != null) {
for (Subsample s : subsamples) {
mzTab.append(s.toMzTab());
}
}
return mzTab.toString();
}
private void parseField(String subId, String field, String value) throws MzTabParsingException {
logger.debug("parsing field: subId = " + subId + ", field = " + field + ", value = " + value);
try {
// simple fields with only one value
if ("title".equals(field)) {
title = value.trim();
} else if ("description".equals(field) && subId == null) {
description = value.trim();
} else if ("false_discovery_rate".equals(field)) {
falseDiscoveryRate = new ParamList(value);
} else if ("uri".equals(field)) {
uri = new URI(value);
} else if ("mod".equals(field)) {
mod = new ParamList(value);
} else if ("quantification_method".equals(field)) {
quantificationMethod = new Param(value);
} else if ("protein-quantification_unit".equals(field)) {
proteinQuantificationUnit = new Param(value);
} else if ("peptide-quantification_unit".equals(field)) {
peptideQuantificationUnit = new Param(value);
}
/** Complex fields with multiple values */
// sample processing
else if (field.startsWith("sample_processing")) {
int sampleProcessingIndex =
Integer.parseInt(
field.substring(18, field.length() - 1)); // extract the processing step number
// create the array if necessary
if (sampleProcessing == null) {
sampleProcessing = new ArrayList<ParamList>();
}
// set the param
sampleProcessing.add(sampleProcessingIndex - 1, new ParamList(value));
} // instruments
else if (field.startsWith("instrument")) {
// get the instrument's index
int instrumentIndex = Integer.parseInt(field.substring(11, field.indexOf(']', 11)));
// create the instrument array if necessary
if (instrument == null) {
instrument = new ArrayList<Instrument>();
}
// create the instrument if necessary
if (instrument.get(instrumentIndex - 1) == null) {
instrument.add(instrumentIndex - 1, new Instrument());
}
// check which value is set
if (field.endsWith("name")) {
instrument.get(instrumentIndex - 1).setName(new Param(value));
} else if (field.endsWith("source")) {
instrument.get(instrumentIndex - 1).setSource(new Param(value));
} else if (field.endsWith("analyzer")) {
instrument.get(instrumentIndex - 1).setAnalyzer(new Param(value));
} else if (field.endsWith("detector")) {
instrument.get(instrumentIndex - 1).setDetector(new Param(value));
}
} // software
else if (field.startsWith("software")) {
// get the software's 1-based index
int softwareIndex = Integer.parseInt(field.substring(9, field.length() - 1));
// create the software array if necessary
if (software == null) {
software = new ArrayList<Param>();
}
// add the software
software.add(softwareIndex - 1, new Param(value));
} // software[1-n]-setting
else if (field.startsWith("software") && field.contains("]-setting")) {
// get the software's 1-based index
int softwareIndex = Integer.parseInt(field.substring(9, field.length() - 1));
// create the software map if necessary
if (softwareSetting == null) {
softwareSetting = new HashMap<Integer, List<String>>();
}
// create the list for this software
if (!softwareSetting.containsKey(softwareIndex)) {
softwareSetting.put(softwareIndex, new ArrayList<String>());
}
// add the setting
softwareSetting.get(softwareIndex).add(value);
} // publication
else if (field.equals("publication")) {
// split the string
String[] publications = value.trim().split("\\|");
// create the publications array if necessary
if (publication == null) {
publication = new ArrayList<String>(publications.length);
}
publication.addAll(Arrays.asList(publications));
} // contact
else if (field.startsWith("contact")) {
// get the instrument's index
int contactIndex = Integer.parseInt(field.substring(8, field.indexOf(']', 8)));
// create the instrument array if necessary
if (contact == null) {
contact = new ArrayList<Contact>();
}
// create the instrument if necessary
if (contact.size() < contactIndex) {
contact.add(contactIndex - 1, new Contact());
}
// check which value is set
if (field.endsWith("name")) {
contact.get(contactIndex - 1).setName(value.trim());
} else if (field.endsWith("email")) {
contact.get(contactIndex - 1).setEmail(value.trim());
} else if (field.endsWith("affiliation")) {
contact.get(contactIndex - 1).setAffiliation(value.trim());
}
} // ms_file
else if (field.startsWith("ms_file")) {
// get the instrument's index
int msFileIndex = Integer.parseInt(field.substring(8, field.indexOf(']', 8)));
// create the instrument array if necessary
if (msFiles == null) {
msFiles = new HashMap<Integer, MsFile>();
}
// create the instrument if necessary
if (msFiles.size() < msFileIndex) {
msFiles.put(msFileIndex - 1, new MsFile());
}
// check which value is set
if (field.endsWith("id_format")) {
msFiles.get(msFileIndex - 1).setIdFormat(new Param(value));
} else if (field.endsWith("format")) {
msFiles.get(msFileIndex - 1).setFormat(new Param(value));
} else if (field.endsWith("location")) {
msFiles.get(msFileIndex - 1).setLocation(value.trim());
}
} // TODO: define how -custom params are handled and react on that
else if (field.equals("custom")) {
if (customParams == null) {
customParams = new ArrayList<Param>();
}
customParams.add(new Param(value));
} // colunits
else if (field.startsWith("colunit-")) {
Matcher matcher = COLUNIT_PATTERN.matcher(value);
if (!matcher.find()) {
throw new MzTabParsingException("Invalid colunit definition: '" + value + "'.");
}
String column = matcher.group(1);
Param unit = new Param(matcher.group(2));
if (field.equals("colunit-protein")) {
if (colunitProtein == null) {
colunitProtein = new HashMap<String, Param>();
}
colunitProtein.put(column, unit);
} else if (field.equals("colunit-peptide")) {
if (colunitPeptide == null) {
colunitPeptide = new HashMap<String, Param>();
}
colunitPeptide.put(column, unit);
} else if (field.equals("colunit-small_molecule")) {
if (colunitSmallMolecule == null) {
colunitSmallMolecule = new HashMap<String, Param>();
}
colunitSmallMolecule.put(column, unit);
}
} // species, tissue, cell type, disease - on the unit level
else if (subId == null && field.startsWith("species")) {
// get the instrument's index
int speciesIndex = Integer.parseInt(field.substring(8, field.length() - 1));
// create the instrument array if necessary
if (species == null) {
species = new ArrayList<Param>();
}
species.add(speciesIndex - 1, new Param(value));
} else if (subId == null && field.startsWith("tissue")) {
// get the instrument's index
int tissueIndex = Integer.parseInt(field.substring(7, field.length() - 1));
// create the instrument array if necessary
if (tissue == null) {
tissue = new ArrayList<Param>();
}
tissue.add(tissueIndex - 1, new Param(value));
} else if (subId == null && field.startsWith("cell_type")) {
// get the instrument's index
int cellTypeIndex = Integer.parseInt(field.substring(10, field.length() - 1));
// create the instrument array if necessary
if (cellType == null) {
cellType = new ArrayList<Param>();
}
cellType.add(cellTypeIndex - 1, new Param(value));
} else if (subId == null && field.startsWith("disease")) {
// get the instrument's index
int diseaseIndex = Integer.parseInt(field.substring(8, field.length() - 1));
// create the instrument array if necessary
if (disease == null) {
disease = new ArrayList<Param>();
}
disease.add(diseaseIndex - 1, new Param(value));
}
/** Parse subsample specific data */
else if (subId != null) {
// extract the index
int subIndex = Integer.parseInt(subId.substring(4, subId.length() - 1));
// make sure the index is greater than 0
// (it should be a 1 based index, and the code will break if this is not the case)
if (subIndex < 1) {
throw new MzTabParsingException("Found sub sample index smaller than 1!");
}
// make sure the subsample array exists
if (subsamples == null) {
subsamples = new ArrayList<Subsample>();
}
// make sure this subsample already exists
if (subsamples.size() < subIndex) {
subsamples.add(subIndex - 1, new Subsample(this.unitId, subIndex));
}
Subsample subsample = subsamples.get(subIndex - 1);
// parse the field
if ("description".equals(field)) {
subsample.setDescription(value.trim());
} else if ("quantification_reagent".equals(field)) {
subsample.setQuantificationReagent(new Param(value));
} else if ("custom".equals(field)) {
if (subsample.getCustomParams() == null) {
subsample.setCustomParams(new ArrayList<Param>(1));
}
subsample.getCustomParams().add(new Param(value));
} else if (field.startsWith("species")) {
// get the instrument's index
int speciesIndex = Integer.parseInt(field.substring(8, field.length() - 1));
// create the instrument array if necessary
if (subsample.getSpecies() == null) {
subsample.setSpecies(new ArrayList<Param>());
}
subsample.getSpecies().add(speciesIndex - 1, new Param(value));
} else if (field.startsWith("tissue")) {
// get the instrument's index
int tissueIndex = Integer.parseInt(field.substring(7, field.length() - 1));
// create the instrument array if necessary
if (subsample.getTissue() == null) {
subsample.setTissue(new ArrayList<Param>());
}
subsample.getTissue().add(tissueIndex - 1, new Param(value));
} else if (field.startsWith("cell_type")) {
// get the instrument's index
int cellTypeIndex = Integer.parseInt(field.substring(10, field.length() - 1));
// create the instrument array if necessary
if (subsample.getCellType() == null) {
subsample.setCellType(new ArrayList<Param>());
}
subsample.getCellType().add(cellTypeIndex - 1, new Param(value));
} else if (field.startsWith("disease")) {
// get the instrument's index
int diseaseIndex = Integer.parseInt(field.substring(8, field.length() - 1));
// create the instrument array if necessary
if (subsample.getDisease() == null) {
subsample.setDisease(new ArrayList<Param>());
}
subsample.getDisease().add(diseaseIndex - 1, new Param(value));
}
} else {
logger.warn("Unknown unit field encountered: " + field);
}
} catch (Exception e) {
throw new MzTabParsingException("Failed to parse mztab metadata field: " + e.getMessage(), e);
}
}