use of db.PolymersDB in project Smiles2Monomers by yoann-dufresne.
the class PolymersTreeWidths method main.
public static void main(String[] args) {
if (args.length < 1) {
System.err.println("Command line : java main.PeptidesTreeWidths <peptidesFile> [outfile]");
System.exit(-1);
}
// Parse arguments
String pepDBname = args[0];
BufferedWriter bw = null;
if (args.length > 1) {
File out = new File(args[1]);
if (out.exists())
out.delete();
try {
bw = new BufferedWriter(new FileWriter(out));
bw.write("peptide;treewidth\n");
} catch (IOException e) {
e.printStackTrace();
}
}
// Loading databases
MonomersDB monoDB = new MonomersDB();
PolymersJsonLoader pcl = new PolymersJsonLoader(monoDB);
PolymersDB pepDB = pcl.loadFile(pepDBname);
// Tools for treewidth
MolTreeWidth mtw = new MolTreeWidth();
for (Polymer pep : pepDB.getObjects()) {
int tw;
try {
tw = mtw.calculateTreeWidth(SmilesConverter.conv.transform(pep.getSmiles()));
} catch (InvalidSmilesException e) {
System.err.println("Impossible to parse " + pep.getName() + " id:" + pep.getId());
continue;
}
if (args.length > 1) {
try {
bw.write(pep.getName() + ";" + tw + "\n");
} catch (IOException e) {
e.printStackTrace();
}
} else {
System.out.println(pep.getName() + " : " + tw);
}
}
if (args.length > 1) {
try {
bw.close();
} catch (IOException e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
}
}
use of db.PolymersDB in project Smiles2Monomers by yoann-dufresne.
the class PreComputation method main.
public static void main(String[] args) {
// ----------------- Parameters ---------------------------
String rulesDBName = "data/rules.json";
String monosDBName = "data/monomers.json";
String jsonPolymers = "data/learning.json";
String serialFolder = "data/serials/";
String jsonResidues = "data/residues.json";
String jsonChains = "data/chains.json";
int markovianSize = 3;
// Parsing
for (int idx = 0; idx < args.length; idx++) {
if (args[idx].startsWith("-")) {
switch(args[idx]) {
case "-rul":
rulesDBName = args[idx + 1];
break;
case "-mono":
monosDBName = args[idx + 1];
break;
case "-poly":
jsonPolymers = args[idx + 1];
break;
case "-res":
jsonResidues = args[idx + 1];
break;
case "-cha":
jsonChains = args[idx + 1];
break;
case "-serial":
serialFolder = args[idx + 1];
break;
case "-markovian":
markovianSize = new Integer(args[idx + 1]);
break;
default:
System.err.println("Wrong option " + args[idx]);
System.exit(1);
break;
}
idx++;
} else {
System.err.println("Wrong parameter " + args[idx]);
System.exit(1);
}
}
// File existence
File file = new File(rulesDBName);
if (!file.exists()) {
System.err.println("No file at " + rulesDBName);
System.exit(1);
}
file = new File(monosDBName);
if (!file.exists()) {
System.err.println("No file at " + monosDBName);
System.exit(1);
}
file = new File(jsonPolymers);
if (!file.exists()) {
System.err.println("No file at " + jsonPolymers);
System.exit(1);
}
// ------------------- Loadings ------------------------
System.out.println("--- Loading ---");
RulesDB rules = RulesJsonLoader.loader.loadFile(rulesDBName);
MonomersDB monos = new MonomersJsonLoader(true).loadFile(monosDBName);
PolymersJsonLoader pjl = new PolymersJsonLoader(monos, false);
PolymersDB learningBase = pjl.loadFile(jsonPolymers);
// ----------------- Serializations --------------------
System.out.println("--- Data serialisation ---");
File folder = new File(serialFolder);
if (!folder.exists())
folder.mkdir();
MonomersSerialization ms = new MonomersSerialization();
ms.serialize(monos, serialFolder + "monos.serial");
// ----------------- residues --------------------------
ResidueCreator rc = new ResidueCreator(rules);
rc.setVerbose(false);
System.out.println("--- Residues creation ---");
FamilyDB families = rc.createResidues(monos);
System.out.println("--- Saving residues ---");
ResidueJsonLoader rjl = new ResidueJsonLoader(rules, monos);
rjl.saveFile(families, jsonResidues);
// ----------------- chains ----------------------------
System.out.println("--- Learning chains ---");
// Adapt residue structures
for (Residue res : families.getResidues().getObjects()) res.explicitToImplicitHydrogens();
ChainLearning learning = new ChainLearning(learningBase);
learning.setMarkovianSize(markovianSize);
learning.learn(families);
// --- Save ---
System.out.println("--- Saving chains ---");
FamilyChainIO fcio = new FamilyChainIO(families);
fcio.saveFile(learning.getDb(), jsonChains);
System.out.println("--- Ended ---");
}
use of db.PolymersDB in project Smiles2Monomers by yoann-dufresne.
the class ProcessPolymers method main.
public static void main(String[] args) {
// ----------------- Parameters ---------------------------
String monoDBname = "data/monomers.json";
String pepDBname = "data/polymers.json";
String rulesDBname = "data/rules.json";
String residuesDBname = "data/residues.json";
String chainsDBFile = "data/chains.json";
String outfile = "results/coverages.json";
String outfolderName = "results/";
String imgsFoldername = "images/";
boolean html = false;
boolean zip = false;
String serialFolder = "data/serials/";
boolean lightMatch = true;
boolean verbose = false;
int removeDistance = 2;
int retryCount = 2;
int modulationDepth = 2;
// Parsing
loop: for (int idx = 0; idx < args.length; idx++) {
if (args[idx].startsWith("-")) {
switch(args[idx]) {
case "-rul":
rulesDBname = args[idx + 1];
break;
case "-mono":
monoDBname = args[idx + 1];
break;
case "-poly":
pepDBname = args[idx + 1];
break;
case "-res":
residuesDBname = args[idx + 1];
break;
case "-cha":
chainsDBFile = args[idx + 1];
break;
case "-serial":
serialFolder = args[idx + 1];
break;
case "-outfile":
outfile = args[idx + 1];
break;
case "-outfolder":
outfolderName = args[idx + 1];
break;
case "-imgs":
imgsFoldername = args[idx + 1];
break;
case "-strict":
lightMatch = false;
continue loop;
case "-v":
verbose = true;
continue loop;
case "-html":
html = true;
continue loop;
case "-zip":
zip = true;
continue loop;
default:
System.err.println("Wrong option " + args[idx]);
System.exit(1);
break;
}
idx++;
} else {
System.err.println("Wrong parameter " + args[idx]);
System.exit(1);
}
}
// ------------------- Loadings ------------------------
System.out.println("--- Loading ---");
// Maybe loading can be faster for the learning base, using serialized molecules instead of CDK SMILES parsing method.
long loadingTime = System.currentTimeMillis();
MonomersDB monoDB = new MonomersJsonLoader(false).loadFile(monoDBname);
MonomersSerialization ms = new MonomersSerialization();
ms.deserialize(monoDB, serialFolder + "monos.serial");
boolean d2 = html || zip;
PolymersJsonLoader pjl = new PolymersJsonLoader(monoDB, d2);
PolymersDB polDB = pjl.loadFile(pepDBname);
RulesDB rulesDB = RulesJsonLoader.loader.loadFile(rulesDBname);
ResidueJsonLoader rjl = new ResidueJsonLoader(rulesDB, monoDB);
// Need optimizations
FamilyDB families = rjl.loadFile(residuesDBname);
FamilyChainIO fcio = new FamilyChainIO(families);
ChainsDB chains = fcio.loadFile(chainsDBFile);
loadingTime = System.currentTimeMillis() - loadingTime;
System.out.println("Loading time : " + (loadingTime / 1000) + "s");
// ------------------- Spliting ------------------------
System.out.println("--- Monomers search ---");
long searchTime = System.currentTimeMillis();
MonomericSpliting.setVerbose(verbose);
MonomericSpliting split = new MonomericSpliting(families, chains, removeDistance, retryCount, modulationDepth);
split.setAllowLightMatchs(lightMatch);
Coverage[] covs = split.computeCoverages(polDB);
searchTime = System.currentTimeMillis() - searchTime;
System.out.println("Search time : " + (searchTime / 1000) + "s");
// ------------------- Output ------------------------
System.out.println("--- Output creations ---");
long outputTime = System.currentTimeMillis();
// Creation of the out directory
File outfolder = new File(outfolderName);
if (!outfolder.exists())
outfolder.mkdir();
CoveragesJsonLoader cjl = new CoveragesJsonLoader(polDB, families);
cjl.saveFile(covs, outfile);
// Images generation
if (html || zip) {
File imgsFolder = new File(imgsFoldername);
if (!imgsFolder.exists())
imgsFolder.mkdir();
ImagesGeneration ig = new ImagesGeneration();
ig.generateMonomerImages(imgsFolder, monoDB);
Map<Coverage, ColorsMap> colors = ig.generatePeptidesImages(imgsFolder, covs);
if (html) {
// HTML
Coverages2HTML c2h = new Coverages2HTML(covs, monoDB, families);
File htmlFile = new File(outfolderName + "/s2m.html");
c2h.createResults(htmlFile, imgsFolder, colors);
}
if (zip) {
// Zip File
OutputZiper oz = new OutputZiper(outfolderName + "/s2m.zip");
oz.createZip(imgsFolder.getPath(), outfile, pepDBname, monoDBname, residuesDBname, colors);
}
}
outputTime = System.currentTimeMillis() - outputTime;
System.out.println("Ouputing time : " + (outputTime / 1000) + "s");
System.out.println("--- Ended ---");
}
use of db.PolymersDB in project Smiles2Monomers by yoann-dufresne.
the class ContractedGraphTests method setUp.
@Before
public void setUp() throws Exception {
MonomersDB monos = new MonomersJsonLoader().loadFile("data_tests/monos.json");
PolymersDB peps = new PolymersJsonLoader(monos).loadFile("data_tests/peps.json");
RulesDB rules = RulesJsonLoader.loader.loadFile("data_tests/rules.json");
ResidueCreator rc = new ResidueCreator(rules);
FamilyDB families = rc.createResidues(monos);
ChainLearning cl = new ChainLearning(peps);
cl.setMarkovianSize(3);
cl.learn(families);
ChainsDB chains = cl.getDb();
MonomericSpliting ms = new MonomericSpliting(families, chains, 2, 2, 3);
Polymer pol = peps.getObject("633");
ms.computeCoverage(pol);
this.coverage = ms.getCoverage();
this.coverage.calculateGreedyCoverage();
this.contractedGraph = new ContractedGraph(coverage);
}
Aggregations