use of algorithms.isomorphism.chains.ChainsDB in project Smiles2Monomers by yoann-dufresne.
the class ChainLearningTests method saveTest.
/**/
@Test
public void saveTest() {
String line = null;
String expected = "[{\"roots\":{\"86\":\"C,0,C,1,2,-1,-1;C,0,N,1,1,0,-1;C,0,C,0,1,-1,0;C,1,O,0,0,3,-1\",\"87\":\"C,0,O,2,0,-1,-1;C,0,C,1,2,-1,0;C,0,C,0,1,-1,2;C,0,N,1,1,2,-1;C,1,O,0,0,3,-1\"},\"family\":\"D-Ser\",\"extensions\":{\"79\":[{\"idx2\":-1,\"idx1\":3,\"from\":\"85\",\"type\":\"extension\",\"ext\":\"C,0,O,0,1\"}],\"78\":[{\"idx2\":-1,\"idx1\":3,\"from\":\"84\",\"type\":\"extension\",\"ext\":\"C,0,O,0,1\"}],\"77\":[{\"num\":1,\"idx\":1,\"from\":\"84\",\"type\":\"hydrogen\"}],\"82\":[{\"idx2\":-1,\"idx1\":3,\"from\":\"87\",\"type\":\"extension\",\"ext\":\"C,0,O,0,1\"}],\"83\":[{\"idx2\":-1,\"idx1\":3,\"from\":\"86\",\"type\":\"extension\",\"ext\":\"C,0,O,0,1\"}],\"80\":[{\"idx2\":-1,\"idx1\":3,\"from\":\"81\",\"type\":\"extension\",\"ext\":\"C,0,O,0,1\"}],\"81\":[{\"num\":1,\"idx\":1,\"from\":\"87\",\"type\":\"hydrogen\"}],\"86\":[],\"87\":[],\"84\":[{\"num\":1,\"idx\":4,\"from\":\"87\",\"type\":\"hydrogen\"}],\"85\":[{\"num\":1,\"idx\":2,\"from\":\"86\",\"type\":\"hydrogen\"}]}},{\"roots\":{\"95\":\"C,0,N,2,1,-1,-1;C,0,C,1,2,-1,0;C,0,N,1,1,2,-1;C,0,C,0,1,-1,2;C,1,O,0,0,4,-1\"},\"family\":\"Dpr\",\"extensions\":{\"95\":[],\"94\":[{\"idx2\":-1,\"idx1\":4,\"from\":\"95\",\"type\":\"extension\",\"ext\":\"C,0,O,0,1\"}],\"93\":[{\"num\":1,\"idx\":3,\"from\":\"95\",\"type\":\"hydrogen\"}],\"92\":[{\"num\":1,\"idx\":1,\"from\":\"95\",\"type\":\"hydrogen\"}],\"91\":[{\"idx2\":-1,\"idx1\":4,\"from\":\"93\",\"type\":\"extension\",\"ext\":\"C,0,O,0,1\"}],\"90\":[{\"idx2\":-1,\"idx1\":4,\"from\":\"92\",\"type\":\"extension\",\"ext\":\"C,0,O,0,1\"}],\"89\":[{\"num\":1,\"idx\":1,\"from\":\"93\",\"type\":\"hydrogen\"}]}}]";
// Computing
ChainLearning learning = new ChainLearning(this.learningBase);
learning.setMarkovianSize(3);
learning.learn(this.families);
ChainsDB db = learning.getDb();
FamilyChainIO io = new FamilyChainIO(this.families);
io.saveFile(db, "data_tests/chains.json");
try {
BufferedReader br = new BufferedReader(new FileReader(new File("data_tests/chains.json")));
line = br.readLine();
br.close();
} catch (IOException e) {
fail(e.getMessage());
}
Assert.assertTrue(line.equals(expected));
}
use of algorithms.isomorphism.chains.ChainsDB 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 algorithms.isomorphism.chains.ChainsDB in project Smiles2Monomers by yoann-dufresne.
the class ChainLearningTests method loadTest.
/**/
@Test
public void loadTest() {
// Computing
ChainLearning learning = new ChainLearning(this.learningBase);
learning.setMarkovianSize(3);
learning.learn(this.families);
ChainsDB db = learning.getDb();
FamilyChainIO io = new FamilyChainIO(this.families);
io.saveFile(db, "data_tests/chains.json");
ChainsDB loaded = io.loadFile("data_tests/chains.json");
if (loaded.getObjects().size() != db.getObjects().size())
fail("Not the same number of objects");
FamilyChainsDB fc = db.getObjects().get(0);
FamilyChainsDB fcLoaded = loaded.getObjects().get(0);
if (fc.getRootChains().size() != fcLoaded.getRootChains().size())
fail("Root chain number different");
for (Residue res : fc.getFamily().getResidues()) {
if (fc.getAdds(res).size() != fcLoaded.getAdds(res).size())
fail("Adds of " + res.getName() + "are not correctly loaded");
}
Assert.assertTrue(true);
}
use of algorithms.isomorphism.chains.ChainsDB 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