DRUG-TARGET INTERACTION PREDICTION

TopoPharmDTI

Integrated Topological and Pharmacophore Feature-Enhanced Deep Learning with Protein Language Model Evaluation for Drug-Target Interactions Prediction

Get Started

Feature representation Dual-tier drug features fusion strategy

Protein language model Model selection strategy

Model performance Notable prediction accuracy

Tested on large scale datasets

The results underscore that our model exhibits superior accuracy and generalization capabilities

120K

Data Scale

50K+

Drug Molecules

5

Datasets

#

Model Workflow

Please provide input

Amino Acid Sequence:

... or, upload a protein file: (csv file, comma separate, column names needed)

SMILES String(s):

... or, upload a compound file:
(csv file, SMILES strings should correspond to amino acid sequences)

If you submit files, the text input will be ignored.

Please wait while computing

Getting started

To specify your desired starting point of the analysis you have to use the input form at the MCPI product page.

You can search MCPI by single amino acid sequence (in any format) and SMILES string. We will return the forecast results of the model based on your input. The value of 0 indicates that the possibility of interaction is low, and the value of 1 indicates that the possibility of interaction is high.
Commonly, you can search by submitting files of proteins and compounds.These files support the fasta and csv file formats.We will calculate each instance in the protein file and compound file you submitted.