extractGInumber

Script

import java.io.File;
import java.io.FileWriter;
import java.util.ArrayList;
import java.util.List;
import java.util.Scanner;
import java.util.regex.Matcher;
import java.util.regex.Pattern;



String output;
output = "";
		
		Scanner getSequence =new Scanner(new File(filename));
		

		
		//assign a pattern only filter the number
		String regEx="[^0-9]";   
		Pattern p = Pattern.compile(regEx);   
		String record = new String();

	//storing all the GI hits inforamtion (100%)
        	List storeList = new ArrayList();
       	//for every GI hit, temporially store the information
	 List tempList = new ArrayList();      
        	//storing the identified GI number (100%)
	List temp123List = new ArrayList();

          //temporary  virable

		String s = getSequence.next();
		String tep = new String();
	// get the GI using the pattern defined
		while(getSequence.hasNext())
		{
			//identify a GI hit where starts
			if(s.startsWith(">"))
			{
				tempList.add(s);
				s=getSequence.next();
				//locate the exact location before the similarity percentage
				while(!getSequence.hasNext("Identities"))
				{
					tempList.add(s);
					s = getSequence.next();			
				}
				tempList.add(s);
				s = getSequence.next();
				tempList.add(s);
				s = getSequence.next();
				tempList.add(s);
				s = getSequence.next();
				tempList.add(s);
				s = getSequence.next();

				String a=s;
				Matcher m = p.matcher(a);   
				if(Integer.parseInt(m.replaceAll("").trim())>98)
				{
					while(!s.startsWith("Query"))
							{
								tempList.add(s);
								s = getSequence.next();
							}
					for(int l =0;l