diff --git a/src/main/java/nl/wur/bif/pantools/pangenome/AnnotationLayer.java b/src/main/java/nl/wur/bif/pantools/pangenome/AnnotationLayer.java
index a85bf6da677bdc637c948d465ea51551ad1ec280..96a4e176e02f1527fd9c97736106a4ba3148502b 100755
--- a/src/main/java/nl/wur/bif/pantools/pangenome/AnnotationLayer.java
+++ b/src/main/java/nl/wur/bif/pantools/pangenome/AnnotationLayer.java
@@ -337,7 +337,9 @@ public class AnnotationLayer {
* @param annotation_file Path to the GFF3 file
* @param proteins_directory The directory to write protein FASTA files in
*/
- private void parse_gff(int genome, String annotation_id, BufferedWriter log_file, String annotation_file, String proteins_directory, Node annotation_node) {
+ private void parse_gff(int genome, String annotation_id, BufferedWriter log_file, String annotation_file,
+ String proteins_directory, Node annotation_node) {
+
int i, trsc, num_genes, num_mRNAs, num_tRNAs, num_rRNAs, feature_len, offset, num_cds_without_genes;
MAX_TRANSACTION_SIZE = 50000; // commit changes every 50k lines
long seq_len = -1, line_nr = 0;
@@ -563,10 +565,16 @@ public class AnnotationLayer {
} // while lines
in.close();
- System.out.println("\r" + address[0] + "\t" + (num_genes + num_cds_without_genes) + "\t" + num_mRNAs + "\t" + num_tRNAs + "\t" + num_rRNAs + "\t");
+ System.out.print("\r" + address[0] + "\t" + (num_genes + num_cds_without_genes) + "\t" + num_mRNAs + "\t"
+ + num_tRNAs + "\t" + num_rRNAs + "\t");
if (num_genes == 0 && num_cds_without_genes == 0) {
- System.out.println("\nNo genes could be added for genome " + address[0] +"!\n");
+ System.out.println("-> No gene nodes could be created!\n");
+ undo_failed_annotation(annotation_node);
+ System.out.println("\rNone of the final GFF's annotations were included to the pangenome\n"
+ + "Please verify the correctness of: " + annotation_file + "\n");
System.exit(1);
+ } else {
+ System.out.println("");
}
try (Transaction tx = graphDb.beginTx()) {
@@ -586,7 +594,54 @@ public class AnnotationLayer {
System.out.println("Could not open " + annotation_file + "!");
}
}
-
+
+ /**
+ * Use in case something went wrong during the annotation and you want to rollback the latest annotation
+ * @param annotation_node an 'annotation' node
+ */
+ private void undo_failed_annotation(Node annotation_node) {
+ int trsc = 0;
+ long node_counter = 0;
+ Transaction tx = graphDb.beginTx(); // start database transaction
+ try {
+ String id = (String) annotation_node.getProperty("identifier");
+ Iterable<Relationship> anno_relations = annotation_node.getRelationships();
+ for (Relationship rel : anno_relations) {
+ rel.delete();
+ }
+ annotation_node.delete();
+ ResourceIterator<Node> annotated_nodes = graphDb.findNodes(feature_label, "annotation_id", id);
+ while (annotated_nodes.hasNext()) {
+ node_counter ++;
+ Node new_node = annotated_nodes.next();
+ Iterable<Relationship> new_relations = new_node.getRelationships();
+ for (Relationship rel : new_relations) {
+ rel.delete();
+ trsc ++;
+ if (trsc >= 100000) { // 100k actions
+ tx.success();
+ tx.close();
+ trsc = 0;
+ tx = graphDb.beginTx(); // start a new database transaction
+ }
+ }
+ trsc ++;
+ if (trsc >= 100000) { // 100k actions
+ tx.success();
+ tx.close();
+ trsc = 0;
+ tx = graphDb.beginTx(); // start a new database transaction
+ }
+ if (node_counter % 5000 == 0){
+ System.out.print("\rRolling back annotated features: " + node_counter);
+ }
+ }
+ } finally {
+ tx.close();
+ }
+ }
+
+
/**
* Parses a GenBank file and annotates the genomes at the same time.
*
@@ -1161,7 +1216,8 @@ public class AnnotationLayer {
create_skip_arrays(false, true); // create skip arrays for sequences and genomes if -skip/-ref is provided by user
String genome_or_identifier = "genome";
- if ((target_genome != null || skip_genomes != null) && PATH_TO_THE_ANNOTATIONS_FILE == null) { // option 1, remove all from selected genomes
+
+ if ((target_genome != null || skip_genomes != null) && PATH_TO_THE_ANNOTATIONS_FILE == null) { // option 1, remove all from selected genomes
try (Transaction tx = graphDb.beginTx()) {
ResourceIterator<Node> annotation_nodes = graphDb.findNodes(annotation_label);
while (annotation_nodes.hasNext()) {
@@ -1225,8 +1281,8 @@ public class AnnotationLayer {
/**
*
- * @param target_label
- * @param identifier
+ * @param target_label a node label
+ * @param identifier an annotation identifier
* @param genome_or_identifier
*/
public void remove_node_rels_matching_genome_or_identifier(Label target_label, String identifier, String genome_or_identifier) {
@@ -1252,8 +1308,9 @@ public class AnnotationLayer {
rel.delete();
trsc1 ++;
rel_counter ++;
- if (rel_counter % 100 == 0 || rel_counter == rel_set.size()) {
- System.out.print("\rRemoving " + target_label + " relationships of annotation " + identifier + ": " + rel_counter + "/" + rel_set.size() + " ");
+ if (rel_counter % 10000 == 0 || rel_counter == rel_set.size()) {
+ System.out.print("\rRemoving " + target_label + " relationships of annotation " + identifier + ": "
+ + rel_counter + "/" + rel_set.size() + " ");
}
if (trsc1 >= MAX_TRANSACTION_SIZE) {
tx.success();
@@ -1275,8 +1332,9 @@ public class AnnotationLayer {
trsc1 ++;
node1.delete();
node_counter ++;
- if (node_counter % 1000 == 0 || node_counter == node_set.size()) {
- System.out.print("\rRemoving " + target_label + " nodes of annotation " + identifier + ": " + node_counter + "/" + node_set.size() + " ");
+ if (node_counter % 10000 == 0 || node_counter == node_set.size()) {
+ System.out.print("\rRemoving " + target_label + " nodes of annotation " + identifier + ": "
+ + node_counter + "/" + node_set.size() + " ");
}
if (trsc1 >= MAX_TRANSACTION_SIZE) {
tx.success();
diff --git a/src/main/java/nl/wur/bif/pantools/pangenome/Classification.java b/src/main/java/nl/wur/bif/pantools/pangenome/Classification.java
index f33513e59fdcf10edc9a5a614a163131b12372c4..56df71d500d732103315c4d8a4acbfe7c7476311 100644
--- a/src/main/java/nl/wur/bif/pantools/pangenome/Classification.java
+++ b/src/main/java/nl/wur/bif/pantools/pangenome/Classification.java
@@ -232,7 +232,7 @@ public class Classification {
ProteomeLayer proLayer = new ProteomeLayer();
StringBuilder output_builder = new StringBuilder();
- System.out.print("\rGathering pangenome statistics");
+ System.out.print("\rGathering pangenome statistics1");
ArrayList<String> phenotype_properties = retrieve_phenotypes_properties();
HashMap<String, String> all_phenotypes_map = retrieve_phenotypes_for_metrics();
StringBuilder genome_csv_builder = create_header_for_genome_metrics_csv_file(phenotype_properties);
@@ -4117,7 +4117,8 @@ public class Classification {
}
/**
- * Key is annotation identifers and also stores the genome number as key. The genome number matches the annotation id from annotation_identifiers
+ * Key is annotation identifers and also stores the genome number as key.
+ * The genome number matches the annotation id from annotation_identifiers
* On a panproteome the returned hashmap is empty as no annotation nodes are present
* @return
*/
@@ -4210,7 +4211,7 @@ public class Classification {
}
/**
- *
+ * Get all the annotation identifiers of annotation nodes
* @return
*/
public static HashMap<Integer, ArrayList<String>> get_all_annotation_identifiers() {
diff --git a/src/main/java/nl/wur/bif/pantools/pangenome/ProteomeLayer.java b/src/main/java/nl/wur/bif/pantools/pangenome/ProteomeLayer.java
index fee24cf2fdb45b5d659c60cc7b72f9e919d1bb83..3b1e31aaffbcb4b7b6ea57ee2960391f2f0537a6 100755
--- a/src/main/java/nl/wur/bif/pantools/pangenome/ProteomeLayer.java
+++ b/src/main/java/nl/wur/bif/pantools/pangenome/ProteomeLayer.java
@@ -2730,8 +2730,8 @@ public class ProteomeLayer {
/**
* Sets 'longest_transcript' property to mRNA, exon, intron and CDS nodes.
*
- * @param count_nodes
- * @param original_number_of_proteins
+ * @param count_nodes count the total number of longest transcripts
+ * @param original_number_of_proteins number of protein
* @return number of proteins, only counting the longest transcripts
*/
public int find_longest_transcript_per_gene(boolean count_nodes, int original_number_of_proteins) {
@@ -2752,17 +2752,17 @@ public class ProteomeLayer {
}
boolean file_exists = check_if_file_exists(WORKING_DIRECTORY + "proteins/longest_transcripts/proteins_" + anno_id + ".fasta");
if (file_exists && !count_nodes) {
- continue;
+ continue; // longest transcripts are already set and counting is not required
}
try (Transaction tx = graphDb.beginTx()) { // start database transaction, commit changes after every genome
String[] anno_id_array = anno_id.split("_");
duplicate_id_builder.append("#Genome ").append(anno_id_array[0]).append("\n");
ResourceIterator<Node> gene_nodes = graphDb.findNodes(gene_label, "annotation_id", anno_id);
- int gene_counter = 0, prot_per_genome = 0, prot_per_genome2 = 0;
+ int gene_counter = 0; // prot_per_genome = 0, prot_per_genome2 = 0;
StringBuilder longest_proteins = new StringBuilder();
while (gene_nodes.hasNext()) {
gene_counter ++;
- if (gene_counter % 100 == 0 || gene_counter == 1) {
+ if (gene_counter % 1000 == 0 || gene_counter == 1) {
System.out.print("\rFinding longest transcript per gene: Genome " + anno_id_array[0] + ", " + gene_counter + " ");
}
Node gene_node = gene_nodes.next();
@@ -2777,7 +2777,7 @@ public class ProteomeLayer {
}
if (first) {
first = false;
- prot_per_genome ++;
+ //prot_per_genome ++;
}
String protein = (String) mrna_node.getProperty("protein_sequence");
if (protein.length() > longest_transcript_length) {
@@ -2825,7 +2825,7 @@ public class ProteomeLayer {
}
protein_counter ++;
- prot_per_genome2 ++;
+ //prot_per_genome2 ++;
String protein_id = (String) longest_transcript_node.getProperty("protein_ID");
String protein_sequence = get_protein_sequence(longest_transcript_node);
protein_sequence = split_seq_in_parts_of_80bp(protein_sequence);