This is the backend API for TIN-X.
TIN-X (Target Importance and Novelty eXplorer) is an interactive visualization tool for illuminating associations between diseases and potential drug targets and is publicly available via https://datascience.unm.edu/tin-x/. TIN-X uses natural language processing to identify disease and protein mentions within PubMed content using previously published tools for named entity recognition (NER) of gene/protein and disease names.
See also the repo TIN-X UI.
- "TIN-X version 3: update with expanded dataset and modernized architecture for enhanced illumination of understudied targets", Vincent T. Metzger, Daniel C. Cannon, Jeremy J. Yang, Stephen L. Mathias, Cristian G. Bologa, Anna Waller, Stephan C. Schürer, Dušica Vidović, Keith J. Kelleher, Timothy K. Sheils, Lars Juhl Jensen, Christophe G. Lambert, Tudor I. Oprea, Jeremy S. Edwards, PeerJ 12:e17470, https://doi.org/10.7717/peerj.17470 (2024).
- "TIN-X: target importance and novelty explorer", Daniel C Cannon, Jeremy J Yang, Stephen L Mathias, Oleg Ursu, Subramani Mani, Anna Waller, Stephan C Schürer, Lars Juhl Jensen, Larry A Sklar, Cristian G Bologa, Tudor I. Oprea, Bioinformatics, Volume 33, Issue 16, 2601–2603, (2017) https://doi.org/10.1093/bioinformatics/btx200
mysql> CREATE USER 'tcrd_read_only'@'%' IDENTIFIED BY 'tcrd_read_only';
Query OK, 0 rows affected (0.00 sec)
mysql> CREATE DATABASE tcrd;
Query OK, 1 row affected (0.00 sec)
mysql> GRANT SELECT ON tcrd.* TO 'tcrd_read_only'@'%';
Query OK, 0 rows affected (0.00 sec)
2. (Skip this step if you have been provided a copy of the database to load.) On a machine with the full tcrd database, you can run the script export_tcrd_subset.sh. This creates a file called tcrd_subset.sql.gz that has all of the tcrd tables needed by the TIN-X REST API.
3. Upload the file tcrd_subset.sql.gz to the same folder where you cloned tinxapi on the local machine / server. Then execute the following commands:
$ gunzip tcrd_subset.sql.gz
$ mysql -u root -p tcrd < import_tcrd_subset.sql
$ python compute_nds_rank.py
- Install docker
sudo apt install docker.io - Build the container
./build.sh - Create
secrets.pyby copyingsecrets.example.pyand then updating details. - Run the container
./run.sh - You can use your IDE like normal and changes will be reflected without rebuilding
Navigate to http://localhost:8000
git clone git@bitbucket.org:iterative-consulting/tinxapi.gitcd tinxapisudo apt-get install virtualenv libmysqlclient-dev python-devvirtualenv envsource env/bin/activatepip install -r requirements.txtcd tinxapipython manage.py migratepython manage.py runserver
Got permission denied while trying to connect to the Docker daemon socket at unix:///var/run/docker.sock
$ sudo usermod -a -G docker $USER
$ exit
Then log back in again.
- Edit your mysql configuration:
sudo vim /etc/mysql/mysql.conf.d/mysqld.cnf - Change
bind-addresstobind-address: 0.0.0.0 - Restart MySQL:
sudo service mysql restart
- Configure docker to run on startup:
sudo systemctl enable docker - Configure tinx container to always run:
docker run -dit --restart unless-stopped -v $PWD:/tinx -p 8000:8000 -ti tinx
You must use solr:6.6.6, Solr8 does not work correctly with Haystack. There is a bug with django-haystack where it does not handle django_ct correctly. There is no fix for this issue presently so we are capping our version of Solr at 6.6.6 as opposed to editing core files.
If you see an issue involving django_ct or django_id, you are using the wrong version of Solr.
It is not clear to me how the managed-schema file was generated, but it is the only one that works presently, so preserve it.
I've authored scripts to redeploy solr once it is built and you can see that work in redeploy-solr.sh.
I've also created a utility script to update solr indices and that can be found in update_solr_index.sh
Haystack won't add all the necessary schema parameters that Solr needs to run through ./manage.py build_solr_schema
Make the following replacement in managed-schema
<!-- <field name="django_ct" type="string" indexed="true" stored="true" multiValued="false"/> -->
<field name="django_ct" type="text_general" indexed="true" stored="true" multiValued="false"/>
Then add this block after the last element
<!-- NRR manual insert start -->
<!-- Lines from origin managed-schema: -->
<fieldType name="pdate" class="solr.DatePointField" docValues="true"/>
<fieldType name="pdates" class="solr.DatePointField" docValues="true" multiValued="true"/>
<fieldType name="pdouble" class="solr.DoublePointField" docValues="true"/>
<fieldType name="pdoubles" class="solr.DoublePointField" docValues="true" multiValued="true"/>
<fieldType name="pfloat" class="solr.FloatPointField" docValues="true"/>
<fieldType name="pfloats" class="solr.FloatPointField" docValues="true" multiValued="true"/>
<fieldType name="pint" class="solr.IntPointField" docValues="true"/>
<fieldType name="pints" class="solr.IntPointField" docValues="true" multiValued="true"/>
<fieldType name="plong" class="solr.LongPointField" docValues="true"/>
<fieldType name="plongs" class="solr.LongPointField" docValues="true" multiValued="true"/>
<!-- NRR manual insert end -->
Solr-8 can now be restarted and the indexes updated through ./manage.py rebuild_index
- Copy
.env.exampleto.env - Run
docker compose up
- You should create a cron job to back up ./db regularly. You can tar the directory (requires stopping containers) or use mysqldump (https://dev.mysql.com/doc/refman/8.3/en/mysqldump.html) along with the mysql container. (ex:
docker compose exec mysql mysqldump [options] > dump.sql)
- Copy
.env.exampleto.env - Edit .env to have connection information for an external database
- Run
docker compose up
-
The docker compose file comes with a Mysql database however, it is recommended to use an external Mysql database. To connect to an external database, copy the .env.example file to .env and edit the variables.
-
If you decide to use the internal database the datastore location can be changed with DATABASE_VOLUME_PATH (default ./db)
-
The env variable BUILD_COMMAND sets what build the UI uses. It can be one of dev|build-staging|build.
-
To deploy production "docker-compose build && docker-compose down && docker-compose up --force-recreate -d"
-
To develop with docker compose, clone the tinx-ui repository to
../tinx-uiand rundocker-compose -f docker-compose-dev.ymlin the tinx-api directory. Both repositories will be linked and changes to tinx-ui populate. (Currently you need to acquire a database dump, this is likely to change)