DrugMiner is an online database to identify whether one protein is druggable or not. Proteins can interact with many biochemical and chemical components. Drugs are some chemicals which involve in many diverse biological processes and they may potentially regulate the functions of thousands of genes and proteins. One major issue in drug studies is to identify which proteins can potentially interact with drugs. The second step in drug-protein studies is prediction of possible interactions between drugs and proteins. Variety of drugs and proteins and various features of drugs as well as those of proteins have limited their identification in laboratories. DrugMiner group, using effective features in drug-protein interaction has established machine learning-based model with high specificity to predict if one protein can interact with drugs or not. By systematically analyzing public high-throughput experimental data retrieved from DrugBank and Uniprot, we have identified novel features that are important in an interaction between drug and protein. These new features as well as other known features have been integrated in machine learning framework for the training of our target prediction model. The coming version of DrugMiner will be established to predict possible interactions between drugs and proteins.
DrugMiner has provided user-friendly approaches to mine druggable proteins within human proteins.
Search textboxes on search page can be used to query the database by [Protein Uniprot ID (e.g. O75051)], [Gene Symbol (e.g. PLXNA2)], and [Protein name (e.g. Plexin-A2)]. The database can also be queried by Proteins’ amino-acid Sequence. (e.g. MEQRRPWPRALEVDSRSVVLLSVVWVLLAPPAAGMPQFS…). It is recommended for users to improve their search result, they perform their query by the ordered search option.
It is possible to perform more query at once, given they are all of the same type in other boxes. Users can click on the beside the boxes to complete their multi-search survey.
Browse page consists of all the predicted proteins as drug target by DrugMiner. Here the proteins are listed according to their UniProt ID. To simple survey on DrugMiner, it is recommended to use search page.
4. Why and How to submit your data?
DrugMiner has provided an opportunity for researchers to check if their own-synthesized proteins are druggable or not. This is an undeniable break which lets researchers to minimize their task in drug discovery and related fields. In such survey, it is possible to decide whether coded proteins by mutated gene can interact with drugs or not.
Users can submit their own-synthesized proteins in DrugMiner in submit page. It is provided to users to submit several proteins at once. Submitted sequences will be analyzed by DrugMiner. The result of this survey will be sent to the users in a week.
5- How to report problems or suggestions?
Please see here to find out how to contact us.
If you make use of the data presented here, please cite the following article in addition to the primary data sources:
Jamali, A. A., Ferdowsi, R., Razaghi, S. Li, J., Safdari, R., Ebrahimie, E., (2016). DrugMiner; Comparative Analysis of Machine Learning Algorithms for Prediction of Potential Druggable Proteins. DRUG DISCOV TODAY.