This a prototype of a data warehouse for storing processed genomic data such as mutation and copy number variation (CNV) data and their annotation information. Currently the prototype models CNV data by gene.
Preliminary benchmarking on query performance has been performed. See /docs/benchmark for details.
This project is built with Spring Boot, Spring Data, andMongoDB.
Running this application requires Maven and MongoDB.
First start up the MongoDB daemon:
mongod
Build the application as follows.
mvn clean package
Then import all test data into MongoDB by executing scripts, execute the following script
sh scripts/import_data.sh
while in the root directory of the project (where pom.xml resides).
Running the JAR package will automatically start tomcat at http://localhost:8080.
java -jar target/magpie1.0.jar
Currently the prototype implements loading and querying of copy number variation data by gene and associated annotation data such as gene, cell line, and study. JSON specifications of these data types can be found in the /docs/data-model directory.
To load sample data to the databases, run scripts in the /scripts/ folder. The import_data.sh script imports all data while others import individual data types.
REST style queries are implemented for CNV data. Suppose the application is configured in Spring Boot to run on localhost and port 8080, then
http://localhost:8080/copynumbergenecentric/all
returns a JSON of all CNV data.
Criteria on fields can be specified as follows
http://localhost:8080/copynumbergenecentric?geneId=1&sampleId=HCT15_LARGE_INTESTINE
To only return certain fields, the following query can be issued
http://localhost:8080/copynumbergenecentric?geneId=1&fields=geneID,copyNumberValue
Sorting is supported by using the following format
http://localhost:8080/copynumbergenecentric?geneId=1&value=lt:0&sort=sampleId:asc
Currently Paging is supported only for retrieving all records as follows
http://localhost:8080/copynumbergenecentric/viewpage?page=1&size=10&sort=sampleId:asc