KD_MultiSucc: incorporating multi-teacher knowledge distillation and word embeddings for cross-species prediction of protein succinylation sites
Bibliographic record
Abstract
Protein succinylation is a vital post-translational modification (PTM) that involves the covalent attachment of a succinyl group (-CO-CH2-CH2-CO-) to the lysine residue of a protein molecule. The mechanism underlying the succinylation process plays a critical role in regulating protein structure, stability, and function, contributing to various biological processes, including metabolism, gene expression, and signal transduction. Succinylation has also been associated with numerous diseases, such as cancer, neurodegenerative disorders, and metabolic syndromes. Due to its important roles, the accurate prediction of succinylation sites is essential for a comprehensive understanding of the mechanisms underlying succinylation. Although research on the identification of protein succinylation sites has been increasing, experimental methods remain time-consuming and costly, underscoring the need for efficient computational approaches. In this study, we present KD_MultiSucc, a model for cross-species prediction of succinylation sites using Multi-Teacher Knowledge Distillation and Word Embedding. The proposed method leverages the strengths of both Knowledge Distillation and Word Embedding techniques to reduce computational complexity while maintaining high accuracy in predicting protein succinylation sites across species. Experimental results demonstrate that the proposed predictor outperforms existing predictors, providing a valuable contribution to PTM research and biomedical applications. To assist readers and researchers, the codes and resources related to this work have been made freely accessible on GitHub at https://github.com/nuinvtnu/KD_MultiSucc/.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.005 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.003 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.003 | 0.004 |
| Insufficient payload (model declined to judge) | 0.004 | 0.003 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".