/**
* Provided only for convenience.
*
* @see #fastaToAfpChain(ProteinSequence, ProteinSequence, Structure, Structure)
* @throws StructureException
*/
public static AFPChain fastaToAfpChain(
SequencePair<Sequence<AminoAcidCompound>, AminoAcidCompound> alignment,
Structure structure1,
Structure structure2)
throws StructureException {
List<AlignedSequence<Sequence<AminoAcidCompound>, AminoAcidCompound>> seqs =
alignment.getAlignedSequences();
StringBuilder sb1 = new StringBuilder();
for (AminoAcidCompound a : seqs.get(0)) {
sb1.append(a.getBase());
}
try {
ProteinSequence seq1 = new ProteinSequence(sb1.toString());
StringBuilder sb2 = new StringBuilder();
for (AminoAcidCompound a : seqs.get(1)) {
sb1.append(a.getBase());
}
ProteinSequence seq2 = new ProteinSequence(sb2.toString());
LinkedHashMap<String, ProteinSequence> map = new LinkedHashMap<String, ProteinSequence>();
map.put(structure1.getName(), seq1);
map.put(structure2.getName(), seq2);
return fastaToAfpChain(map, structure1, structure2);
} catch (CompoundNotFoundException e) {
logger.error(
"Unexpected error while creating protein sequences: {}. This is most likely a bug.",
e.getMessage());
return null;
}
}
/**
* Print an alignment map in a concise representation. Edges are given as two numbers separated by
* '>'. They are chained together where possible, or separated by spaces where disjoint or
* branched.
*
* <p>Note that more concise representations may be possible. Examples:
* <li>1>2>3>1
* <li>1>2>3>2 4>3
*
* @param alignment The input function, as a map (see {@link
* AlignmentTools#alignmentAsMap(AFPChain)})
* @param identity An identity-like function providing the isomorphism between the codomain of
* alignment (of type <T>) and the domain (type <S>).
* @return
*/
public static <S, T> String toConciseAlignmentString(Map<S, T> alignment, Map<T, S> identity) {
// Clone input to prevent changes
Map<S, T> alig = new HashMap<S, T>(alignment);
// Generate inverse alignment
Map<S, List<S>> inverse = new HashMap<S, List<S>>();
for (Entry<S, T> e : alig.entrySet()) {
S val = identity.get(e.getValue());
if (inverse.containsKey(val)) {
List<S> l = inverse.get(val);
l.add(e.getKey());
} else {
List<S> l = new ArrayList<S>();
l.add(e.getKey());
inverse.put(val, l);
}
}
StringBuilder str = new StringBuilder();
while (!alig.isEmpty()) {
// Pick an edge and work upstream to a root or cycle
S seedNode = alig.keySet().iterator().next();
S node = seedNode;
if (inverse.containsKey(seedNode)) {
node = inverse.get(seedNode).iterator().next();
while (node != seedNode && inverse.containsKey(node)) {
node = inverse.get(node).iterator().next();
}
}
// Now work downstream, deleting edges as we go
seedNode = node;
str.append(node);
while (alig.containsKey(node)) {
S lastNode = node;
node = identity.get(alig.get(lastNode));
// Output
str.append('>');
str.append(node);
// Remove edge
alig.remove(lastNode);
List<S> inv = inverse.get(node);
if (inv.size() > 1) {
inv.remove(node);
} else {
inverse.remove(node);
}
}
if (!alig.isEmpty()) {
str.append(' ');
}
}
return str.toString();
}