private static void CheckFPSmartsForMolecule(string moleculeSmiles, string[][] expectedFPSmarts)
{
var expected = new HashSet <string>();
foreach (var strs in expectedFPSmarts)
{
foreach (var str in strs)
{
expected.Add(str);
}
}
// expectedFPSmarts[][] is a double array because for each smarts
// several equivalent variants
// of the smarts are given e.g. CCC C(C)C
var mol = parser.ParseSmiles(moleculeSmiles);
CircularFingerprinter circ = new CircularFingerprinter();
circ.Calculate(mol);
var subsmarts = new SmartsFragmentExtractor(mol);
subsmarts.SetMode(SubstructureSelectionMode.JCompoundMapper);
var numFP = circ.FPCount;
var actual = new HashSet <string>();
for (int i = 0; i < numFP; i++)
{
var fp = circ.GetFP(i);
actual.Add(subsmarts.Generate(fp.Atoms));
}
Assert.IsTrue(expected.IsSupersetOf(actual));
}
public static string CircularFpToString(CircularFingerprinter cfp)
{
CircularFingerprint fp = null;
string s = "fp\thashCode\titeration\tatoms\r\n";
int fpCount = cfp.FPCount;
for (int fpi = 0; fpi < fpCount; fpi++)
{
fp = cfp.GetFP(fpi);
s += fpi.ToString() + "\t" + fp.Hash + "\t" + fp.Iteration + "\t(" + string.Join(", ", fp.Atoms) + ")\r\n";
}
return(s);
}
/// <summary>
/// Appends a new row to the model source data, which consists of a molecule and whether or not it
/// is considered active.
/// </summary>
/// <param name="mol">molecular structure, which must be non-blank</param>
/// <param name="active">whether active or not</param>
public void AddMolecule(IAtomContainer mol, bool active)
{
if (mol == null || mol.Atoms.Count == 0)
{
throw new CDKException("Molecule cannot be blank or null.");
}
var circ = new CircularFingerprinter(ClassType)
{
PerceiveStereo = this.PerceiveStereo
};
circ.Calculate(mol);
// gather all of the (folded) fingerprints into a sorted set
int AND_BITS = Folding - 1; // e.g. 1024/0x400 -> 1023/0x3FF: chop off higher order bits
var hashset = new SortedSet <int>();
for (int n = circ.FPCount - 1; n >= 0; n--)
{
int code = circ.GetFP(n).Hash;
if (Folding > 0)
{
code &= AND_BITS;
}
hashset.Add(code);
}
// convert the set into a sorted primitive array
var hashes = new int[hashset.Count];
int p = 0;
foreach (var h in hashset)
{
hashes[p++] = h;
}
// record the processed information for model building purposes
if (active)
{
numActive++;
}
training.Add(hashes);
activity.Add(active);
foreach (var h in hashes)
{
if (!inHash.TryGetValue(h, out int[] stash))
/// <summary>
/// BuildTest
/// </summary>
public static void BuildTest()
{
CircularFingerprinter cfp = null;
CircularFingerprinterClass FpClass = CircularFingerprinterClass.ECFP6; // FP diameter
int FpLen = 2048; // folded binary fp length
IAtomContainer mol, mol2;
//string molfile = FileUtil.ReadFile(@"C:\Download\CorpId-12345.mol");
//java.io.StringReader sr = new java.io.StringReader(molfile);
//if (Lex.Contains(molfile, "v2000"))
// cor = new MDLV2000Reader(sr);
//else
// cor = new MDLV3000Reader(sr);
//cor.setReaderMode(IChemObjectReader.Mode.RELAXED);
//ac = (IAtomContainer)cor.read(new AtomContainer());
//cor.close();
FpClass = CircularFingerprinterClass.ECFP4; // debug
cfp = new CircularFingerprinter(FpClass, FpLen);
StreamReader reader = new StreamReader(@"C:\Download\CorpId-12345.mol");
//FileReader FileReader = new FileReader(@"C:\Download\V3000 Mols.sdf");
EnumerableSDFReader rdr = new EnumerableSDFReader(reader.BaseStream, ChemObjectBuilder.Instance);
rdr.ReaderMode = ChemObjectReaderMode.Relaxed;
IEnumerator <IAtomContainer> cursor = rdr.GetEnumerator();
while (cursor.MoveNext())
{
mol = cursor.Current;
mol = CdkMol.GetLargestMoleculeFragment(mol);
ICountFingerprint cfp1 = cfp.GetCountFingerprint(mol); // get hash values and counts for each
cfp.Calculate(mol);
int fpCount = cfp.FPCount;
for (int fpi = 0; fpi < fpCount; fpi++) // gets
{
CircularFingerprint cfp2 = cfp.GetFP(fpi); // gets hash, iteration and lists of atoms (dups appear multiple times)
}
IBitFingerprint bfp = cfp.GetBitFingerprint(mol);
BitArray bs = bfp.AsBitSet();
int bsCard = bfp.Cardinality;
long bsSize = bfp.Length;
continue;
}
reader.Close();
return;
//java.io.StringReader sr = new java.io.StringReader(molfile);
//AtomContainer mol = new AtomContainer();
//mol.addAtom(new Atom("C"));
//mol.addAtom(new Atom("H"));
//mol.addAtom(new Atom("H"));
//mol.addAtom(new Atom("H"));
//mol.addAtom(new Atom("H"));
//mol.addBond(new Bond(mol.getAtom(0), mol.getAtom(1)));
//mol.addBond(new Bond(mol.getAtom(0), mol.getAtom(2)));
//mol.addBond(new Bond(mol.getAtom(0), mol.getAtom(3)));
//mol.addBond(new Bond(mol.getAtom(0), mol.getAtom(4)));
//FileReader FileReader = new FileReader(@"C:\Download\CorpId-12345.mol");
//MolReader mr = new MolReader(FileReader, DefaultChemObjectBuilder.getInstance());
//java.io.StringReader sr = new java.io.StringReader(molfile);
//IMol m = (IMol)mr.next();
//FileReader.close();
}
