/**
* Returns an AFPChain corresponding to the alignment between {@code structure1} and {@code
* structure2}, which is given by the gapped protein sequences {@code sequence1} and {@code
* sequence2}. The sequences need not correspond to the entire structures, since local alignment
* is performed to match the sequences to structures. Assumes that a residue is aligned if and
* only if it is given by an uppercase letter.
*
* @param sequence1 <em>Must</em> have {@link ProteinSequence#getUserCollection()} set to document
* upper- and lower-case as aligned and unaligned; see {@link
* #getAlignedUserCollection(String)}
* @throws StructureException
*/
public static AFPChain fastaToAfpChain(
ProteinSequence sequence1,
ProteinSequence sequence2,
Structure structure1,
Structure structure2)
throws StructureException {
if (structure1 == null || structure2 == null) {
throw new IllegalArgumentException("A structure is null");
}
if (sequence1 == null || sequence2 == null) {
throw new IllegalArgumentException("A sequence is null");
}
Atom[] ca1 = StructureTools.getRepresentativeAtomArray(structure1);
Atom[] ca2 = StructureTools.getRepresentativeAtomArray(structure2);
ResidueNumber[] residues1 =
StructureSequenceMatcher.matchSequenceToStructure(sequence1, structure1);
ResidueNumber[] residues2 =
StructureSequenceMatcher.matchSequenceToStructure(sequence2, structure2);
// nullify ResidueNumbers that have a lowercase sequence character
if (sequence1.getUserCollection() != null) {
CasePreservingProteinSequenceCreator.setLowercaseToNull(sequence1, residues1);
}
if (sequence2.getUserCollection() != null) {
CasePreservingProteinSequenceCreator.setLowercaseToNull(sequence2, residues2);
}
return buildAlignment(ca1, ca2, residues1, residues2);
}
/**
* Takes a structure and sequence corresponding to an alignment between a structure or sequence
* and itself (or even a structure with a sequence), where the result has a circular permutation
* site {@link cpSite} residues to the right.
*
* @param first The unpermuted sequence
* @param second The sequence permuted by cpSite
* @param cpSite The number of residues from the beginning of the sequence at which the circular
* permutation site occurs; can be positive or negative; values greater than the length of the
* sequence are acceptable
* @throws StructureException
*/
public static AFPChain cpFastaToAfpChain(
ProteinSequence first, ProteinSequence second, Structure structure, int cpSite)
throws StructureException {
if (structure == null) {
throw new IllegalArgumentException("The structure is null");
}
if (first == null) {
throw new IllegalArgumentException("The sequence is null");
}
// we need to find the ungapped CP site
int gappedCpShift = 0;
int ungappedCpShift = 0;
while (ungappedCpShift < Math.abs(cpSite)) {
char c;
try {
if (cpSite <= 0) {
c = second.getSequenceAsString().charAt(gappedCpShift);
} else {
c = second.getSequenceAsString().charAt(first.getLength() - 1 - gappedCpShift);
}
} catch (StringIndexOutOfBoundsException e) {
throw new IllegalArgumentException("CP site of " + cpSite + " is wrong");
}
if (c != '-') {
ungappedCpShift++;
}
gappedCpShift++;
}
Atom[] ca1 = StructureTools.getRepresentativeAtomArray(structure);
Atom[] ca2 =
StructureTools.getRepresentativeAtomArray(
structure); // can't use cloneCAArray because it doesn't set parent
// group.chain.structure
ProteinSequence antipermuted = null;
try {
antipermuted =
new ProteinSequence(
SequenceTools.permuteCyclic(second.getSequenceAsString(), gappedCpShift));
} catch (CompoundNotFoundException e) {
// this can't happen, the original sequence comes from a ProteinSequence
logger.error(
"Unexpected error while creating protein sequence: {}. This is most likely a bug.",
e.getMessage());
}
ResidueNumber[] residues = StructureSequenceMatcher.matchSequenceToStructure(first, structure);
ResidueNumber[] antipermutedResidues =
StructureSequenceMatcher.matchSequenceToStructure(antipermuted, structure);
ResidueNumber[] nonpermutedResidues = new ResidueNumber[antipermutedResidues.length];
SequenceTools.permuteCyclic(antipermutedResidues, nonpermutedResidues, -gappedCpShift);
// nullify ResidueNumbers that have a lowercase sequence character
if (first.getUserCollection() != null) {
CasePreservingProteinSequenceCreator.setLowercaseToNull(first, residues);
}
if (second.getUserCollection() != null) {
CasePreservingProteinSequenceCreator.setLowercaseToNull(second, nonpermutedResidues);
}
// for (int i = 0; i < residues.length; i++) {
// if (residues[i] == null) {
// System.out.print("=");
// } else {
// System.out.print(sequence.getSequenceAsString().charAt(i));
// }
// }
// System.out.println();
// for (int i = 0; i < residues.length; i++) {
// if (nonpermutedResidues[i] == null) {
// System.out.print("=");
// } else {
// System.out.print(second.getSequenceAsString().charAt(i));
// }
// }
// System.out.println();
return buildAlignment(ca1, ca2, residues, nonpermutedResidues);
}