Some changes in the validator

OmicsDI · Sep 29, 2016 · 55fb32b · 55fb32b
1 parent 5b0a581
commit 55fb32b
Show file tree

Hide file tree

Showing 6 changed files with 113 additions and 18 deletions.
diff --git a/README.md b/README.md
@@ -63,6 +63,6 @@ file, the command should be refined to:
 
 ## Support
 
-For any issue or support questions, please contact pride-support@ebi.ac.uk
+For any issue or support questions, please contact omicsdi-support@ebi.ac.uk
 
 
diff --git a/src/main/java/uk/ac/ebi/ddi/xml/validator/cli/validatorCLI.java b/src/main/java/uk/ac/ebi/ddi/xml/validator/cli/validatorCLI.java
@@ -89,11 +89,8 @@ public static void main(String[] args) throws Exception {
 
             Map<File, List<Tuple>> errors = new HashMap<>();
             for(File file: files){
-
                 List<Tuple> error = OmicsXMLFile.validateSchema(file);
-                if(checkValue.equalsIgnoreCase(Utils.WARN)){
-                    error.addAll(OmicsXMLFile.validateSemantic(file));
-                }
+                error.addAll(OmicsXMLFile.validateSemantic(file));
                 if(errors.containsKey(file)){
                    error.addAll(errors.get(file));
                 }
@@ -103,18 +100,20 @@ public static void main(String[] args) throws Exception {
             if(!errors.isEmpty()){
                 if(reportName != null){
                     PrintStream reportFile = new PrintStream(new File(reportName));
-                    generateSummaryReport(errors,reportFile);
-                    for(File file: errors.keySet()){
+                    generateSummaryReport(errors,reportFile, checkValue);
+                    for(File file: errors.keySet())
                         for (Tuple error: errors.get(file))
-                            reportFile.println(file.getAbsolutePath() + "\t" + error.getKey() + "\t" + error.getValue());
-                    }
+                            if(checkValue == Utils.WARN || (error.getValue() == Utils.ERROR))
+                                reportFile.println(file.getAbsolutePath() + "\t" + error.getKey() + "\t" + error.getValue());
+
                     reportFile.close();
                 }else{
                     for(File file: errors.keySet()){
                         PrintStream reportFile = new PrintStream(new File(file.getAbsolutePath() + ".error.csv"));
-                        generateSummaryReport(errors,reportFile);
+                        generateSummaryReport(errors,reportFile, checkValue);
                         for (Tuple error: errors.get(file))
-                            reportFile.println(file.getName() + "\t" + error.getKey() + "\t" + error.getValue());
+                            if(checkValue == Utils.WARN || (error.getValue() == Utils.ERROR))
+                                reportFile.println(file.getName() + "\t" + error.getKey() + "\t" + error.getValue());
                         reportFile.close();
                     }
                 }
@@ -123,7 +122,7 @@ public static void main(String[] args) throws Exception {
         }
     }
 
-    public static void generateSummaryReport(Map<File, List<Tuple>> errors, PrintStream reportFile){
+    public static void generateSummaryReport(Map<File, List<Tuple>> errors, PrintStream reportFile, String errorLevel){
         int numberErrors = 0;
         int numberWars   = 0;
         Map<Field, Integer> fields = new HashMap<>();
@@ -150,10 +149,13 @@ public static void generateSummaryReport(Map<File, List<Tuple>> errors, PrintStr
             Field field = (Field) entry.getKey();
             Integer errorNumber = (Integer) entry.getValue();
             String error = (field.getType() == FieldType.MANDATORY)? Utils.ERROR:Utils.WARN;
-            if(field.getCategory() == FieldCategory.DATE)
-                reportFile.println(error + " The number of datasets without or outdated " + field.getFullName() + " is " + errorNumber);
-            else
-                reportFile.println(error + " The number of datasets without " + field.getFullName() + " is " + errorNumber);
+            if(errorLevel == Utils.WARN || (error == Utils.ERROR)){
+                if(field.getCategory() == FieldCategory.DATE)
+                    reportFile.println(error + " The number of datasets without or outdated " + field.getFullName() + " is " + errorNumber);
+                else
+                    reportFile.println(error + " The number of datasets without " + field.getFullName() + " is " + errorNumber);
+
+            }
         }
     }
 

diff --git a/src/main/java/uk/ac/ebi/ddi/xml/validator/utils/Field.java b/src/main/java/uk/ac/ebi/ddi/xml/validator/utils/Field.java
@@ -44,7 +44,8 @@ public enum Field {
     SECONDARY_ACCESSION("additional_accession", FieldType.OPTIONAL,FieldCategory.ADDITIONAL, "Secondary Accession"),
     ENSEMBL_EXPRESSION_ATLAS("ensembl", FieldType.UNKNOWN, FieldCategory.CROSSREF, "Gene reference to ENSEMBL"),
 
-    SUBMITTER_KEYWORDS("submitter_keywords", FieldType.UNKNOWN, FieldCategory.ADDITIONAL, "MODEL");
+    SUBMITTER_KEYWORDS("submitter_keywords", FieldType.UNKNOWN, FieldCategory.ADDITIONAL, "Submitter Keywords"),
+    CURATOR_KEYWORDS("curator_keywords", FieldType.UNKNOWN, FieldCategory.ADDITIONAL, "Submitter Keywords");
 
 
 

diff --git a/src/main/java/uk/ac/ebi/ddi/xml/validator/utils/Utils.java b/src/main/java/uk/ac/ebi/ddi/xml/validator/utils/Utils.java
@@ -3,7 +3,6 @@
 import org.apache.log4j.Logger;
 import uk.ac.ebi.ddi.xml.validator.parser.model.*;
 import uk.ac.ebi.ddi.xml.validator.parser.model.Date;
-import uk.ac.ebi.ddi.xml.validator.parser.unmarshaller.OmicsUnmarshallerFactory;
 
 import java.text.ParseException;
 import java.text.SimpleDateFormat;

diff --git a/src/test/resources/BIOMODELS.xml b/src/test/resources/BIOMODELS.xml
@@ -0,0 +1,74 @@
+<database>
+    <name>BioModels Database</name>
+    <description>BioModels Database is a repository of computational models of biological processes. Models described
+        from literature are manually curated and enriched with cross-references.
+    </description>
+    <release>1</release>
+    <release_date>2016-08-26</release_date>
+    <entry_count>1</entry_count>
+    <entries>
+        <entry id="MODEL00002243.1"> <!-- uniqueID -->
+            <name>Thiele2013 - Cerebellum cells in molecular layer</name>
+            <description>
+                Thiele2013 - Cerebellum cells in molecular layer The model of cerebellum cells in molecular layer
+                metabolism is derived from the community-driven global reconstruction of human metabolism
+                (version 2.02, MODEL1109130000 ).
+            </description>
+            <dates>
+                <date type="submission" value="2016-08-19"/>
+                <date type="publication" value="2016-08-20"/>
+                <date type="last_modification" value="2016-08-19"/>
+            </dates>
+            <additional_fields>
+                <field name="submitter">Tung Nguyen</field>
+                <field name="submitter_mail">tnguyen@ebi.ac.uk</field>
+                <field name="submitter_affiliation">EMBL - European Bioinformatics Institute</field>
+                <field name="repository">BioModels</field>
+                <field name="full_dataset_link">http://wwwdev.ebi.ac.uk/biomodels/jummp-biomodels/model/MODEL00002243.1</field>
+                <!--<field name="publication">Pub Title + Pub Authors + Pub Abstract + Pub Year</field>-->
+                <field name="publication">A community-driven global reconstruction of human metabolism,
+                    Thiele I, et al . (2013)
+                    Multiple models of human metabolism have been reconstructed, but each represents only a subset of
+                    our knowledge. Here we describe Recon 2, a community-driven, consensus 'metabolic reconstruction',
+                    which is the most comprehensive representation of human metabolism that is applicable to
+                    computational modeling. Compared with its predecessors, the reconstruction has improved topological
+                    and functional features, including ~2x more reactions and ~1.7x more unique metabolites. Using
+                    Recon 2 we predicted changes in metabolite biomarkers for 49 inborn errors of metabolism with 77%
+                    accuracy when compared to experimental data. Mapping metabolomic data and drug information onto
+                    Recon 2 demonstrates its potential for integrating and analyzing diverse data types. Using protein
+                    expression data, we automatically generated a compendium of 65 cell type-specific models, providing
+                    a basis for manual curation or investigation of cell-specific metabolic properties. Recon 2 will
+                    facilitate many future biomedical studies and is freely available at http://humanmetabolism.org/.
+                </field>
+                <field name="disease">Disease name</field>
+                <!--<field name="omics_type">Models</field>-->
+                <field name="data_protocol">extracted from Curation comments</field> <!-- Software used -->
+                <field name="sample_protocol">extracted from Curation comments</field> <!-- Software used -->
+                <field name="technology_type">Software used</field> <!-- Software used -->
+
+                <!-- The additional fields we define ourself -->
+                <field name="modelFormat">SBML</field>
+                <field name="submissionId">UNCHECKED</field>
+                <field name="publicationId">UNCHECKED</field>
+                <field name="levelVersion">UNCHECKED</field>
+                <field name="validationStatus">UNCHECKED</field>
+                <field name="certificationComment">This model is not certified</field>
+                <field name="elementName">Element Name</field>
+                <field name="elementID">Element ID</field>
+                <field name="elementDescription">Element Description</field>
+                <field name="public">The model is public</field>
+                <field name="deleted">The model is deleted</field>
+                <field name="certified">The model is certified</field>
+                <field name="sbmlSBOTerm">SBML SBO Term</field>
+                <field name="curators">Curators</field>
+                <field name="authors">Authors</field>
+                <field name="derivations">Derivations</field>
+                <field name="pharmml**">Fields related to PharmML</field>
+            </additional_fields>
+            <cross_references>
+                <ref dbkey="CHEBI:16551" dbname="ChEBI"/>
+                <ref dbkey="MTBLC30031" dbname="MetaboLights"/>
+            </cross_references>
+        </entry>
+    </entries>
+</database>
diff --git a/src/test/resources/BIOMODELS.xml.error.csv b/src/test/resources/BIOMODELS.xml.error.csv
@@ -0,0 +1,19 @@
+The number of Errors in the files: 1
+The number of Warnings in the files: 7
+Warn The number of datasets without Dataset NCBI TAXONOMY is 1
+Warn The number of datasets without Dataset Submitter email is 1
+Warn The number of datasets without MODEL is 8
+Warn The number of datasets without Instrument Platform is 1
+Warn The number of datasets without Gene reference to ENSEMBL is 1
+Warn The number of datasets without Dataset Pubmed Id is 1
+Warn The number of datasets without Secondary Accession is 1
+Error The number of datasets without Dataset Omics Type is 1
+Warn The number of datasets without MEDLINE Reference is 1
+BIOMODELS.xml	Warn	Entry: MODEL00002243.1 The entry do not contain: Dataset NCBI TAXONOMY
+BIOMODELS.xml	Warn	Entry: MODEL00002243.1 The entry do not contain: Dataset Pubmed Id
+BIOMODELS.xml	Warn	Entry: MODEL00002243.1 The entry do not contain: MEDLINE Reference
+BIOMODELS.xml	Warn	Entry: MODEL00002243.1 The entry do not contain: Gene reference to ENSEMBL
+BIOMODELS.xml	Error	Entry: MODEL00002243.1 The entry do not contain: Dataset Omics Type
+BIOMODELS.xml	Warn	Entry: MODEL00002243.1 The entry do not contain: Dataset Submitter email
+BIOMODELS.xml	Warn	Entry: MODEL00002243.1 The entry do not contain: Instrument Platform
+BIOMODELS.xml	Warn	Entry: MODEL00002243.1 The entry do not contain: Secondary Accession