From physics to prediction: genetic algorithm-optimized neural network using hansen solubility parameters for pharmaceutical solubility in neat and mixed solvents
Bibliographic record
Abstract
Reliable prediction of active pharmaceutical ingredient (API) solubility in complex solvent systems remains challenging. Existing models often sacrifice generalizability for accuracy or have limited applicability due to implementation complexity. The current study presents a practical, streamlined modeling framework requiring only temperature, solvent composition, and Hansen solubility parameters (HSPs) — information that is simple and readily accessible. These parameters serve as inputs for an effective and easy-to-apply model: the multilayer perceptron artificial neural network (MLPANN). Beyond optimizing the network architecture with a genetic algorithm (GA), model accuracy is further supported by six input scenarios designed to explore alternative HSP formulations and dimensional reduction strategies. To ensure generality, the MLPANN was trained on 496 experimental solubility values of acetaminophen, diazepam, ibuprofen, lorazepam, and naproxen in both neat and binary solvent systems, including water, ethanol, isopropanol, dioxane, NMP, and propylene glycol. Two scenarios, using the optimized networks, achieved R 2 values exceeding 0.99 across training, validation, and testing subsets. Graphical validation—an important aspect often overlooked in previous studies—demonstrated excellent predictive performance for unseen testing data of acetaminophen in isopropanol–water and naproxen in ethanol–water mixtures.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 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".