The ActiveDriverDB database contains hundreds of thousands of genome variants that affect protein signalling sites of post-translational modifications (PTMs). Using this information, we can predict functions of genome variation and disease mutations on protein signalling networks. The database is available at activedriverdb.org and provides visual browsing, application programming interfaces and data analysis of custom uploaded datasets of mutations.
The database integrates nearly 400 000 experimentally verified PTM sites and about 3.5 million of amino-acid substitutions from cancer genome sequencing studies (TCGA, The Cancer Genome Atlas), carefully curated disease mutations from the ClinVar database (ClinVar) as well as population genome sequencing studies (ESP6500 and 1000 Genomes Project).
We cover four types of PTMs that have the most abundant experimental data about human proteins: phosphorylation involved in many key cellular and developmental pathways, ubiquitination primarily involved in protein degradation, and acetylation and methylation involved in epigenetic and transcriptional regulation networks.
Visualizations and views
The ActiveDriverDB provides two main ways to explore and interpret the mutations in context of PTM sites: sequence and network views.
The sequence view consists of a “needle plot” which shows how often a specific position in protein’s sequence is mutated (especially in PTM sites; shown in coloured circles). An interactive table and adaptive set of filters enable easy customization of the view. The sequence view also shows the impact of mutations on sequence motifs bound by kinases and predicts gains and losses of motifs.
The network view shows how kinases interact with PTM sites of the chosen protein and the predicted effect of mutation on motifs in the sequence. Two variations of the network view are available: experimental, which presents known protein site-kinase interactions, and MIMP-predicted, which highlights interactions that are predicted to occur based on the sequence motifs and included motifs mutations occur. Additionally, kinase-targeting drugs from DrugBank are presented in the network view.
How to start?
Researchers interested in particular genes, pathways or diseases may use the search bar on the front page to start exploring how known mutations might influence the post-translational modification sites of their genes and proteins of interest.
If you wish to test which mutations from your experiment affect or might affect PTM sites, please use advanced mutation search form: https://activedriverdb.org/search/mutations.
To explore the most interesting genes with mutations in PTM sites, please see one of the available lists: activedriverdb.org/gene/lists/. For alike list of pathways, visit: activedriverdb.org/pathways/.
For more information, please see our online tutorial: https://activedriverdb.org/tutorial/, which briefly explains sequence and network views, navigation, search functionality and and other features of the website.
Advanced users might want to access network, mutations or protein data using their programming language of choice; to that end the access with REST API is supported, with the documentation available at https://activedriverdb.org/api/.
Methods, data sources and case studies are all explained in detail in our primary publication in NAR: https://academic.oup.com/nar/article/46/D1/D901/4566599.
Michal Krassowski, Marta Paczkowska, Kim Cullion, Tina Huang, Irakli Dzneladze, B. F. Francis Ouellette, Joseph T. Yamada, Amelie Fradet-Turcotte, Jüri Reimand. ActiveDriverDB: human disease mutations and genome variation in post-translational modification sites of proteins. Nucleic Acids Research 2018, gkx973, https://doi.org/10.1093/nar/gkx973
The database is open-source and the source code is available on GitHub: github.com/reimandlab/ActiveDriverDB.