Protein–Protein Interactions Prediction via Multimodal Deep Polynomial Network and Regularized Extreme Learning Machine
Problem
Predicting which proteins interact with which others, a key task in understanding cellular processes, using computational methods instead of costly experiments.
Why it matters
Experimentally mapping protein-protein interactions is slow and expensive, so accurate computational prediction accelerates biology and drug research.
What the AI did
A multimodal deep polynomial network integrated multiple protein features in two stages, and a regularized extreme learning machine performed the final PPI prediction.
Results
not stated in source
Impact
The method was tested on a public dataset, but the abstract gives no performance numbers.
Record
| Venue | IEEE Journal of Biomedical and Health Informatics |
|---|---|
| Published | 2018 |
| Work type | article |
| Authors | Haijun Lei, Yuting Wen, Zhu‐Hong You, Ahmed Elazab, Ee-Leng Tan, Yujia Zhao, Baiying Lei |
| Citations | 42 (OpenAlex, at import time) |
| References | 84 |
| Model or method | Multimodal deep polynomial network (MDPN) with regularized extreme learning machine |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1109/jbhi.2018.2845866 |
| Open access | green |
| Keywords | protein-protein interactions, deep polynomial network, extreme learning machine, multimodal, prediction |
Importance rating
real but narrower advance
PPI prediction supports biology and drug research applications per abstract, with high reported accuracies, but no scale of impact is stated; significant but not field-defining evidence.