demo.CookbookMSA Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of biojava-alignment Show documentation
The newest version!
/**
 *                    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/
 *
 * Created on Jul 2, 2012
 * Created by Andreas Prlic
 *
 * @since 3.0.2
 */
package demo;

import org.biojava.nbio.alignment.Alignments;
import org.biojava.nbio.core.alignment.template.Profile;
import org.biojava.nbio.core.sequence.ProteinSequence;
import org.biojava.nbio.core.sequence.compound.AminoAcidCompound;
import org.biojava.nbio.core.sequence.io.FastaReaderHelper;
import org.biojava.nbio.core.util.ConcurrencyTools;

import java.net.URL;
import java.util.ArrayList;
import java.util.List;

public class CookbookMSA {

	public static void main(String[] args) throws Exception {
		String[] ids = new String[] {"Q21691", "A8WS47", "O48771"};
		multipleSequenceAlignment(ids);
	}

	private static void multipleSequenceAlignment(String[] ids) throws Exception {
		List lst = new ArrayList<>();
		for (String id : ids) {
			lst.add(getSequenceForId(id));
		}
		Profile profile = Alignments.getMultipleSequenceAlignment(lst);
		System.out.printf("Clustalw:%s%s", System.getProperty("line.separator"), profile);
		System.out.println();

		ConcurrencyTools.shutdown();
	}

	private static ProteinSequence getSequenceForId(String uniProtId) throws Exception {
		URL uniprotFasta = new URL(String.format("http://www.uniprot.org/uniprot/%s.fasta", uniProtId));
		System.out.println("Getting Sequence from URL: "+ uniprotFasta);

		ProteinSequence seq = FastaReaderHelper.readFastaProteinSequence(uniprotFasta.openStream()).get(uniProtId);
		System.out.printf("id : %s %s%s%s", uniProtId, seq, System.getProperty("line.separator"), seq.getOriginalHeader());
		System.out.println();

		return seq;
	}

}