/*
 *                    BioJava development code
 *
 * This code may be freely distributed and modified under the
 * terms of the GNU Lesser General Public Licence.  This should
 * be distributed with the code.  If you do not have a copy,
 * see:
 *
 *      http://www.gnu.org/copyleft/lesser.html
 *
 * Copyright for this code is held jointly by the individual
 * authors.  These should be listed in @author doc comments.
 *
 * For more information on the BioJava project and its aims,
 * or to join the biojava-l mailing list, visit the home page
 * at:
 *
 *      http://www.biojava.org/
 *
 */
package org.biojava.nbio.alignment;

import org.biojava.nbio.core.alignment.matrices.SimpleSubstitutionMatrix;
import org.biojava.nbio.alignment.Alignments.PairwiseSequenceAlignerType;
import org.biojava.nbio.core.alignment.template.SequencePair;
import org.biojava.nbio.core.alignment.template.SubstitutionMatrix;
import org.biojava.nbio.core.sequence.ProteinSequence;
import org.biojava.nbio.core.sequence.compound.AminoAcidCompound;
import org.biojava.nbio.core.sequence.io.FastaReaderHelper;

import java.net.URL;

public class CookbookAlignPairGlobal {

	public static void main(String[] args) {
		String[] ids = new String[] {"Q21691", "Q21495", "O48771"};
		try {
			alignPairGlobal(ids[0], ids[1]);
		} catch (Exception e){
			e.printStackTrace();
		}
	}

	private static void alignPairGlobal(String id1, String id2) throws Exception {
		ProteinSequence s1 = getSequenceForId(id1), s2 = getSequenceForId(id2);
		SubstitutionMatrix<AminoAcidCompound> matrix = SimpleSubstitutionMatrix.getBlosum62();
		SequencePair<ProteinSequence, AminoAcidCompound> pair = Alignments.getPairwiseAlignment(s1, s2,
				PairwiseSequenceAlignerType.GLOBAL, new SimpleGapPenalty(), matrix);
		System.out.printf("%n%s vs %s%n%s", pair.getQuery().getAccession(), pair.getTarget().getAccession(), pair);
	}

	private static ProteinSequence getSequenceForId(String uniProtId) throws Exception {
		URL uniprotFasta = new URL(String.format("http://www.uniprot.org/uniprot/%s.fasta", uniProtId));
		ProteinSequence seq = FastaReaderHelper.readFastaProteinSequence(uniprotFasta.openStream()).get(uniProtId);
		System.out.printf("id : %s %s%n%s%n", uniProtId, seq, seq.getOriginalHeader());
		return seq;
	}

}