Predicting biological activity from biosynthetic gene clusters using neural networks
Bibliographic record
Abstract
Abstract Microorganisms like bacteria and fungi have been used for natural products that translate to drugs. However, assessing the bioactivity of extract from culture to identify novel natural molecules remains a strenuous process due to the cumbersome order of production, purification, and assaying. Thus, extensive genome mining of microbiomes is underway to identify biosynthetic gene clusters or BGCs that can be profiled as particular natural products, and computational methods have been developed to address this problem using machine learning. However, existing tools are ineffective due to a small training dataset, dependence on old genome mining tools, lack of relevant genomic descriptors, and prevalent class imbalance. This work presents a new tool, NPBdetect, that can detect multiple bioactivities and has been designed through rigorous experiments. Firstly, we composed a larger training set using MIBiG database and a test set through literature mining to build and assess the model respectively. Secondly, the latest antiSMASH genome mining tool was used to obtain BGC and introduced new sequence-based descriptors. Thirdly, neural networks are used to build the model by dealing with class imbalance issues through the class weighting technique. Finally, we compared the NPBdetect tool with an existing tool to show its efficacy and real-world utility in detecting several bioactivities with high confidence.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 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".