Add journal metadata table for recording impact factor.

.gitignore

@@ -30,6 +30,6 @@ assets/*
 .VSCodeCounter
 neo4j-data
 neo4j-import
-data/drugbank/*
-data/aact_*/*
+data/entity_attr/drugbank/*
+data/entity_attr/aact_*/*
 frontend

data/README.md

@@ -1,34 +0,0 @@
-## Requirements
-
-- Python 3.11+
-- json2parquet
-
-```bash
-# macOSx
-brew install domoritz/homebrew-tap/json2parquet
-
-# Linux
-cargo install json2parquet
-```
-
-- Required Python packages: `pip install click duckdb`
-
-## Prepare additional data for each entity and relation
-
-### Compound
-
-Get additional data for each compound from [DrugBank](https://www.drugbank.ca/). You might need to request access to the DrugBank data. If you have access, download the DrugBank XML file and save it to the `data` directory. We assume the file is named `drugbank_5.1_2024-01-03.xml`.
-
-```bash
-# Convert the DrugBank XML file to TSV
-python3 data/drugbank.py tojson --input-file data/drugbank/drugbank_5.1_2024-01-03.xml --output-dir data/drugbank --format tsv && zip data/drugbank/drugbank_5.1_2024-01-03.tsv.zip data/drugbank/drugbank_5.1_2024-01-03.tsv
-
-# Convert the DrugBank XML file to JSON
-python3 data/drugbank.py tojson --input-file data/drugbank/drugbank_5.1_2024-01-03.xml --output-dir data/drugbank
-
-# Convert the JSON file to Parquet
-python3 data/drugbank.py tojson --input-file data/drugbank/drugbank_5.1_2024-01-03.xml --output-dir data/drugbank --format linejson
-json2parquet data/drugbank/drugbank_5.1_2024-01-03.jsonl data/drugbank/drugbank_5.1_2024-01-03.parquet
-```
-
-### Gene

data/entity_attr/README.md

@@ -0,0 +1,34 @@
+## Requirements
+
+- Python 3.11+
+- json2parquet
+
+```bash
+# macOSx
+brew install domoritz/homebrew-tap/json2parquet
+
+# Linux
+cargo install json2parquet
+```
+
+- Required Python packages: `pip install click duckdb`
+
+## Prepare additional data for each entity and relation
+
+### Compound
+
+Get additional data for each compound from [DrugBank](https://www.drugbank.ca/). You might need to request access to the DrugBank data. If you have access, download the DrugBank XML file and save it to the `data` directory. We assume the file is named `drugbank_5.1_2024-01-03.xml`.
+
+```bash
+# Convert the DrugBank XML file to TSV
+python3 data/drugbank.py tojson --input-file data/drugbank/drugbank_5.1_2024-01-03.xml --output-dir data/drugbank --format tsv && zip data/drugbank/drugbank_5.1_2024-01-03.tsv.zip data/drugbank/drugbank_5.1_2024-01-03.tsv
+
+# Convert the DrugBank XML file to JSON
+python3 data/drugbank.py tojson --input-file data/drugbank/drugbank_5.1_2024-01-03.xml --output-dir data/drugbank
+
+# Convert the JSON file to Parquet
+python3 data/drugbank.py tojson --input-file data/drugbank/drugbank_5.1_2024-01-03.xml --output-dir data/drugbank --format linejson
+json2parquet data/drugbank/drugbank_5.1_2024-01-03.jsonl data/drugbank/drugbank_5.1_2024-01-03.parquet
+```
+
+### Gene

migrations/20240709_add_journal_metadata_table.down.sql

@@ -0,0 +1,2 @@
+-- Drop the journal metadata table when rolling back the migration.
+DROP TABLE IF EXISTS biomedgps_journal_metadata;

migrations/20240709_add_journal_metadata_table.up.sql

@@ -0,0 +1,18 @@
+-- biomedgps_journal_metadata table is created to store metadata for journals, such as the journal name, the journal type, etc.
+
+CREATE TABLE IF NOT EXISTS biomedgps_journal_metadata (
+    journal_name VARCHAR(255) NOT NULL UNIQUE, -- The name of the journal
+    abbr_name VARCHAR(255) NOT NULL UNIQUE, -- The abbreviation name of the journal
+    issn VARCHAR(32) NOT NULL UNIQUE, -- The print ISSN of the journal
+    eissn VARCHAR(32) NOT NULL UNIQUE, -- The electronic ISSN of the journal
+    impact_factor DECIMAL(6, 3), -- The impact factor of the journal
+    impact_factor_5_year DECIMAL(6, 3), -- The 5-year impact factor of the journal
+    category VARCHAR(32), -- The category of the journal, such as Medicine, Biology, etc.
+    jcr_quartile VARCHAR(8), -- Journal Citation Reports (JCR) quartile, such as Q1, Q2, etc.
+    rank INTEGER, -- The rank of the journal in the category
+    total_num_of_journals INTEGER, -- The total number of journals in the category
+    CONSTRAINT biomedgps_journal_metadata_journal_name_uniq_key UNIQUE (journal_name),
+    CONSTRAINT biomedgps_journal_metadata_abbr_name_uniq_key UNIQUE (abbr_name),
+    CONSTRAINT biomedgps_journal_metadata_issn_uniq_key UNIQUE (issn),
+    CONSTRAINT biomedgps_journal_metadata_eissn_uniq_key UNIQUE (eissn)
+);

src/api/route.rs

@@ -12,8 +12,8 @@ use crate::model::core::{
     Entity, Entity2D, EntityMetadata, KnowledgeCuration, RecordResponse, Relation, RelationCount,
     RelationMetadata, Statistics, Subgraph,
 };
-use crate::model::entity_attr::EntityAttr;
-use crate::model::entity_attr::{CompoundAttr, EntityAttrRecordResponse};
+use crate::model::entity::compound::CompoundAttr;
+use crate::model::entity_attr::{EntityAttr, EntityAttrRecordResponse};
 use crate::model::graph::Graph;
 use crate::model::init_db::get_kg_score_table_name;
 use crate::model::kge::DEFAULT_MODEL_NAME;

src/bin/biomedgps-cli.rs

@@ -1,6 +1,6 @@
 extern crate log;
 
-use biomedgps::model::entity_attr::CompoundAttr;
+use biomedgps::model::entity::compound::CompoundAttr;
 use biomedgps::model::init_db::create_kg_score_table;
 use biomedgps::model::kge::{init_kge_models, DEFAULT_MODEL_NAME};
 use biomedgps::model::{
@@ -11,7 +11,8 @@ use biomedgps::model::{
     util::read_annotation_file,
 };
 use biomedgps::{
-    build_index, connect_graph_db, import_data, import_kge, init_logger, run_migrations, change_emb_dimension
+    build_index, change_emb_dimension, connect_graph_db, import_data, import_kge, init_logger,
+    run_migrations,
 };
 use log::*;
 use regex::Regex;
@@ -700,7 +701,13 @@ async fn main() {
             if table_name == "biomedgps" && arguments.dimension != 400 {
                 if arguments.force {
                     warn!("The dimension of the embedding is not 400, but the table name is biomedgps. We will change the dimension of the embedding as you specified.");
-                    match change_emb_dimension(&database_url, table_name.as_str(), arguments.dimension).await {
+                    match change_emb_dimension(
+                        &database_url,
+                        table_name.as_str(),
+                        arguments.dimension,
+                    )
+                    .await
+                    {
                         Ok(_) => {
                             info!("Change the dimension of the embedding successfully.");
                         }