\n ![](\"../assets/drift-net-logo.png\")
\n
\n \n \n \n \n \n | User ID: | |
| User Name: | |
| Email: |
\n Off-target binding is one of the primary causes of toxic side effects of drugs in clinical development, resulting in failures of clinical trials. While off-target drug binding is a known phenomenon, experimental identification of the undesired protein binders is prohibitively expensive due to a large pool of possible biological targets. Here, we propose a new strategy based on the \"chemical similarity principle\" that enables proteome-wide identification of off-target protein binders of chemical compounds based on a vast and expanding knowledge of protein ligand complexes.
\nWe develop a methodology to generate and compare path-based binary fingerprints for both query and annotated compounds and formulate computational algorithms to quantify the chemical similarity between the query and the annotated bait compounds to infer targets. We also build a user-friendly web server for drug-target identification based on chemical similarity (DRIFT) to perform searches across annotated bioactive compound datasets, thus enabling high-throughput, multi-ligand target identification, and chemical fragmentation.
\n\nWe provide a user-friendly interface to submit a query compound. Users can either search the compound, upload the compound file, or draw the compound structure.
\n\n![](\"../assets/submit-query-compound.gif\")
![](\"../assets/upload-query-compound.gif\")
![](\"../assets/draw-query-compound.gif\")
Database: Please choose either \"CHEMBL\", \"HMDB (Metabolites)\", \"ZINC\", or \"BindingDB\".
\n\nSimilarity: CutoffThe default similarity cutoff is 0.85.
\n\nNumber: The number of compounds that would selected for further targets identification. The default number if 100.
\n\nTo support submitting multiple compounds at one time, we provide the batch mode.
\n![](\"../assets/batch-mode-example.gif\")
Please feel free to contact us if you have more questions.
\nPlease upload the SMILES file:
\nor input the SMILES of all compounds:
\n![](\"../assets/draw-structure-logo.png\"){kind=logo}
![](\"../assets/upload-structure-logo.png\"){kind=logo}
Drift is a computational tool to predict protein targets for any unknown chemical compound.
\nDrift generates FP2 binary fingerprint (path-based) for query compound and performs substructure search against large curated datasets of bioactive molecules with drug-like properties (ChEMBL, ZINC, and HMDB). The protein targets of identified bait compounds will be ranked and listed based on the extent of similarity with the query structure. Thus, Drift methodology expedite off-target prediction and drug development pipeline for any lead compound identified from experimental assays.
\n \n![](\"../assets/drift-method.png\")
\n \n We are constantly working on improving interfaces and identification accuracy. Please check back often!
\n| Task ID | \nUser | \nTitle | \nSubmit Time | \nProcess Time | \nFinish Time | \nStatus | \n \n \n
|---|---|---|---|---|---|---|
| {{ task.id }} | \n{{`${task.userid}`==='0' ? 'Guest' : task.userid}} | \n{{'title' in task ? task.title : ''}} | \n\n | \n | \n | \n Queued \n Failed \n \n Done \n \n \n \n | \n