initial commit

.classpath

@@ -0,0 +1,9 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<classpath>
+	<classpathentry kind="src" path="src"/>
+	<classpathentry kind="con" path="org.eclipse.jdt.launching.JRE_CONTAINER/org.eclipse.jdt.internal.debug.ui.launcher.StandardVMType/JavaSE-1.8"/>
+	<classpathentry kind="lib" path="C:/lib/commons-codec-1.11/commons-codec-1.11.jar"/>
+	<classpathentry kind="lib" path="C:/lib/ftp4j-1.7.2/ftp4j-1.7.2.jar"/>
+	<classpathentry kind="lib" path="C:/lib/commons-io-2.6/commons-io-2.6.jar"/>
+	<classpathentry kind="output" path="bin"/>
+</classpath>

.project

@@ -0,0 +1,17 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<projectDescription>
+	<name>ENAuploader</name>
+	<comment></comment>
+	<projects>
+	</projects>
+	<buildSpec>
+		<buildCommand>
+			<name>org.eclipse.jdt.core.javabuilder</name>
+			<arguments>
+			</arguments>
+		</buildCommand>
+	</buildSpec>
+	<natures>
+		<nature>org.eclipse.jdt.core.javanature</nature>
+	</natures>
+</projectDescription>

.settings/org.eclipse.jdt.core.prefs

@@ -0,0 +1,11 @@
+eclipse.preferences.version=1
+org.eclipse.jdt.core.compiler.codegen.inlineJsrBytecode=enabled
+org.eclipse.jdt.core.compiler.codegen.targetPlatform=1.8
+org.eclipse.jdt.core.compiler.codegen.unusedLocal=preserve
+org.eclipse.jdt.core.compiler.compliance=1.8
+org.eclipse.jdt.core.compiler.debug.lineNumber=generate
+org.eclipse.jdt.core.compiler.debug.localVariable=generate
+org.eclipse.jdt.core.compiler.debug.sourceFile=generate
+org.eclipse.jdt.core.compiler.problem.assertIdentifier=error
+org.eclipse.jdt.core.compiler.problem.enumIdentifier=error
+org.eclipse.jdt.core.compiler.source=1.8

src/eu/europa/ecdc/enauploader/DatabaseEntity.java

@@ -0,0 +1,52 @@
+package eu.europa.ecdc.enauploader;
+import java.io.File;
+
+public abstract class DatabaseEntity {
+
+	String title;
+	String accession;
+	String centerName;
+	String alias;
+	File xmlFile;
+	String type;
+
+	DatabaseEntity(String c, String a) {
+		title = a;
+		accession = "";
+		centerName=c;
+		alias=a;
+
+	}
+
+	public void setAccession(String a) {
+		accession = a;
+	}
+	public String getAccession() {
+		return accession;
+	}
+
+	public void writeXml() {
+		xmlFile = new File(ENAUtils.TMP_PATH+type+"_"+alias+".xml");
+		writeXml(xmlFile);
+	}
+
+	public abstract void writeXml(File f);
+
+	public abstract String getSubmitRow();
+
+	public Object getXmlFile() {
+		return xmlFile;
+	}
+
+	public String getType() {
+		return type;
+	}
+
+	public String getAlias() {
+		return alias;
+	}
+
+
+
+
+}

src/eu/europa/ecdc/enauploader/ENAUtils.java

@@ -0,0 +1,41 @@
+package eu.europa.ecdc.enauploader;
+import java.io.File;
+import java.io.IOException;
+import java.util.ArrayList;
+import java.util.HashMap;
+
+import it.sauronsoftware.ftp4j.FTPAbortedException;
+import it.sauronsoftware.ftp4j.FTPClient;
+import it.sauronsoftware.ftp4j.FTPDataTransferException;
+import it.sauronsoftware.ftp4j.FTPException;
+import it.sauronsoftware.ftp4j.FTPIllegalReplyException;
+
+public class ENAUtils {
+
+	public static final String FTP_HOST = "webin.ebi.ac.uk";
+	public static final String TMP_PATH = "C:/ENAtmp/";
+	public static final String CURL_PATH = "T:/Epidemiological Methods/MolecularSurveillance/Software/curl.exe";
+
+
+	public static void init() {
+		File tmpDir = new File(TMP_PATH);
+		tmpDir.mkdirs();
+	}
+
+
+
+	public static String getTaxid(String taxon) {
+
+		switch (taxon) {
+
+		case "Listeria monocytogenes":
+			return "1639";
+		case "Salmonella enterica":
+			return "28901";
+		case "Escherichia coli":
+			return "562";
+		default:
+			return "0";
+		}
+	}
+}

src/eu/europa/ecdc/enauploader/ENAuploaderMain.java

@@ -0,0 +1,103 @@
+package eu.europa.ecdc.enauploader;
+import java.io.File;
+import java.text.SimpleDateFormat;
+import java.util.ArrayList;
+import java.util.Date;
+import java.util.UUID;
+
+public class ENAuploaderMain {
+
+	// This is a command line version
+	public static void main(String[] args) {
+		//args: 
+		//Center
+		//Project id ("" means create new project) "PRJEB25143"
+		//CSV file path
+		//Data directory path
+		// Login (Webin-NNNNN)
+		// Password
+		// yes/no whether to use the production database at EBI
+
+		if (args.length < 4) {
+			return;
+		}
+
+		String center = args[0];
+		String project = args[1];
+		File csvFile = new File(args[2]);
+		File dataDir = new File(args[3]);
+
+		String login = "";
+		String pass = "";
+		String prod = "no";
+		if (args.length > 5) {
+			login = args[4];
+			pass = args[5];
+		}
+
+		if (args.length > 6) {
+			prod = args[6];
+		}
+
+		SubmissionFactory.createSubmissionFromCSV(center,project, csvFile,dataDir, login, pass, prod);
+	}
+
+
+	//Example code for use of the library, rename to main to run it
+	public static void main2(String[] args) {
+
+		//Init tmp folder
+		ENAUtils.init();
+
+		//base name for testing
+		String randomUUIDString = UUID.randomUUID().toString();
+		String sname = "ECDCtest_"+randomUUIDString;
+
+		//Create new submission
+		Submission s = new Submission("ECDC","sub_"+sname);
+		//s.SetLogin("login", "pwd"); //Use this to change credentials from ECDC test account
+		//s.useProductionServer(false); //Use this to change between test and production on EBI 
+		//(Test submissions are deleted within 24 hours)
+
+		//New project
+		//Project p = new Project("ECDC","proj_"+sname);
+		//s.addEntry(p);
+
+
+		//New sample wrapper (includes sample, experiment and run)
+		SampleWrapper wrap = new SampleWrapper("ECDC","PRJEB25143", sname);
+
+		//Set minimal metadata
+		wrap.sample.setAttribute("host_associated","yes");
+		wrap.sample.setAttribute("specific_host","human");
+		wrap.sample.setAttribute("host_disease_status","diseased");
+		wrap.sample.setAttribute("collected_by","ECDC");
+		wrap.sample.setAttribute("collection_date","2018");
+		wrap.sample.setAttribute("country","Sweden");
+		wrap.sample.setAttribute("serovar","unknown");
+		wrap.sample.setTaxon("Listeria monocytogenes");
+		wrap.experiment.setInstrument("Ion Torrent S5 XL");
+
+		//Add a sequence file
+		wrap.run.addFile(new File("C:/ENAtmp/raw/MI-90_18b.fastq.gz"));
+
+		//Add entries in wrapper to submission
+		s.addEntry(wrap);
+
+		//Upload files and submit
+		s.uploadFiles();
+		s.submit();
+
+
+		//Test output
+		//System.out.println(p.getAccession());
+		System.out.println(wrap.run.getAccession());
+		System.out.println(wrap.sample.getAccession());
+		System.out.println(wrap.experiment.getAccession());
+		for (File f : wrap.run.getOriginalFiles()) {
+			System.out.println(f.toString());
+		}
+
+	}
+
+}

src/eu/europa/ecdc/enauploader/Experiment.java

@@ -0,0 +1,92 @@
+package eu.europa.ecdc.enauploader;
+import java.io.BufferedWriter;
+import java.io.File;
+import java.io.FileWriter;
+import java.io.IOException;
+
+public class Experiment extends DatabaseEntity {
+
+	private String studyAccession;
+	private String sampleAccession;
+
+	private String pairedDistance;
+	private String pairedDistanceSdev;
+
+	private String instrument;
+
+	Experiment(String c, String a, String studyAcc, String sampleAcc) {
+		super(c, a);
+
+		pairedDistance = "500";
+		pairedDistanceSdev = "200.0";
+
+
+		type = "EXPERIMENT";
+		studyAccession = studyAcc;
+		sampleAccession = sampleAcc;
+	}
+
+	public void setInstrument(String instr) {
+		instrument = instr;
+	}
+
+	@Override
+	public void writeXml(File f) {
+		BufferedWriter bw;
+		try {
+			bw = new BufferedWriter(new FileWriter(f));
+			bw.write("<?xml version = '1.0' encoding = 'UTF-8'?>\n");
+			bw.write("<EXPERIMENT_SET xmlns:xsi=\"http://www.w3.org/2001/XMLSchema-instance\" xsi:noNamespaceSchemaLocation=\"ftp://ftp.sra.ebi.ac.uk/meta/xsd/sra_1_5/SRA.experiment.xsd\">\n");
+			bw.write("<EXPERIMENT alias=\""+alias+"\" center_name=\""+centerName+"\">\n");
+			bw.write("<TITLE>"+title+"</TITLE>\n");
+			if (studyAccession.startsWith("PRJ")) {
+				bw.write("<STUDY_REF accession=\""+studyAccession+"\"/>\n");
+			} else {
+				bw.write("<STUDY_REF refname=\""+studyAccession+"\"/>\n");
+			}
+
+			bw.write("<DESIGN>\n");
+			bw.write("<DESIGN_DESCRIPTION>Whole genome sequencing of pathogens</DESIGN_DESCRIPTION>\n");
+			bw.write("<SAMPLE_DESCRIPTOR refname=\""+sampleAccession+"\"/>\n");
+			bw.write("<LIBRARY_DESCRIPTOR>\n");
+			bw.write("<LIBRARY_NAME>"+alias+"</LIBRARY_NAME>\n");
+			bw.write("<LIBRARY_STRATEGY>WGS</LIBRARY_STRATEGY>\n");
+			bw.write("<LIBRARY_SOURCE>GENOMIC</LIBRARY_SOURCE>\n");
+			bw.write("<LIBRARY_SELECTION>RANDOM</LIBRARY_SELECTION>\n");
+			bw.write("<LIBRARY_LAYOUT>\n");
+			if (instrument.matches(".*Illumina.*")) {
+				bw.write("<PAIRED NOMINAL_LENGTH=\""+pairedDistance+"\" NOMINAL_SDEV=\""+pairedDistanceSdev+"\"/>\n");
+			} else {
+				bw.write("<SINGLE/>\n");
+			}
+			bw.write("</LIBRARY_LAYOUT>\n");
+			bw.write("<LIBRARY_CONSTRUCTION_PROTOCOL>unknown</LIBRARY_CONSTRUCTION_PROTOCOL>\n");
+			bw.write("</LIBRARY_DESCRIPTOR>\n");
+			bw.write("</DESIGN>\n");
+			bw.write("<PLATFORM>\n");
+			if (instrument.matches(".*Illumina.*")) {
+				bw.write("<ILLUMINA>\n");
+				bw.write("<INSTRUMENT_MODEL>"+instrument+"</INSTRUMENT_MODEL>\n");
+				bw.write("</ILLUMINA>\n");	
+			} else if(instrument.matches(".*Ion Torrent.*")) {
+				bw.write("<ION_TORRENT>\n");
+				bw.write("<INSTRUMENT_MODEL>"+instrument+"</INSTRUMENT_MODEL>\n");
+				bw.write("</ION_TORRENT>\n");	
+			}
+			bw.write("</PLATFORM>\n");
+			bw.write("</EXPERIMENT>\n");
+			bw.write("</EXPERIMENT_SET>\n");
+			bw.close();
+		} catch (IOException e) {
+
+			e.printStackTrace();
+		}
+	}
+
+
+	@Override
+	public String getSubmitRow() {
+		return "<ADD schema=\"experiment\" source=\""+xmlFile.toString()+"\"/>";
+	}
+
+}

src/eu/europa/ecdc/enauploader/Project.java

@@ -0,0 +1,45 @@
+package eu.europa.ecdc.enauploader;
+import java.io.BufferedWriter;
+import java.io.File;
+import java.io.FileWriter;
+import java.io.IOException;
+
+public class Project extends DatabaseEntity {
+
+
+	String description;
+
+
+	Project(String c, String a) {
+		super(c,a);
+		description = a;
+		type = "PROJECT";
+	}
+
+	public void writeXml(File projectFile) {
+
+		BufferedWriter bw;
+		try {
+			bw = new BufferedWriter(new FileWriter(projectFile));
+			bw.write("<?xml version = '1.0' encoding = 'UTF-8'?>\n");
+			bw.write("<PROJECT_SET>\n");
+			bw.write("<PROJECT alias=\""+alias+"\" center_name=\""+centerName+"\">\n");
+			bw.write("<TITLE>"+title+"</TITLE>\n");
+			bw.write("<DESCRIPTION>"+description+"</DESCRIPTION>\n");
+			bw.write("<SUBMISSION_PROJECT>\n");
+			bw.write("<SEQUENCING_PROJECT/>\n");
+			bw.write("</SUBMISSION_PROJECT>\n");
+			bw.write("</PROJECT>\n");
+			bw.write("</PROJECT_SET>\n");
+			bw.close();
+		} catch (IOException e) {
+			// TODO Auto-generated catch block
+			e.printStackTrace();
+		}
+	}
+
+	public String getSubmitRow() {
+		return "<ADD schema=\"project\" source=\""+xmlFile.toString()+"\"/>";
+	}
+
+}