The Use of DeepQSAR Models for the Discovery of Peptides With Enhanced Antimicrobial and Antibiofilm Potential
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Increasing concerns regarding prolonged antibiotic usage have spurred the search for alternative treatments. Antimicrobial peptides (AMPs), first discovered in the 1980s, have exhibited significant potential against a broad range of bacteria. Short-sequenced AMPs are abundant in nature and present across various organisms. Recently, machine learning technologies such as Quantitative Structure Activity Relationships (QSAR) have enabled expedited discovery of potential AMPs with broad-spectrum antibacterial activity as the amount of available AMP training data increases. Among those, Deep QSAR has recently emerged as a distinct type of application that utilizes conventional molecular descriptors in conjunction with more powerful deep learning (DL) models. Here, we demonstrate the power of Deep QSAR in predicting broad-spectrum AMP activity. Using a recurrent neural network-based QSAR model, we achieved nearly 90% fivefold cross-validated accuracy in classifying AMP activity. Using the developed approach, we designed 98 novel peptides, of which 36 experimentally demonstrated more effective antibiofilm activity and 26 peptides exhibited stronger antimicrobial activity compared to a well-characterized host defense peptide IDR-1018, which was demonstrated to possess broad spectrum antibiofilm activity against a wide range of bacterial pathogens and a previous computer-aided peptide design study employing IDR-1018 derivatives successfully identified novel peptides with enhanced antibiofilm activity. Notably, 22 of those peptides demonstrated improvements of both antimicrobial and, particularly, antibiofilm properties, making them suitable prototypes for preclinical development and demonstrating efficacy of DeepQSAR modeling in identifying novel and potent AMPs.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 it