Novel Pharmacokinetics Profiler (PhaKinPro): Model Development, Validation, and Implementation as a Web-Tool for Triaging Compounds with Undesired PK Profiles
Bibliographic record
Abstract
Computational models that predict PK properties, such as those related to drug absorption, metabolism, distribution, and excretion, are critical to flagging drug candidates with poor PK profiles that emerge as hits in high-throughput screening campaigns. To support the development of reliable computational models to predict key PK properties, we collected, curated, and integrated a database of compounds tested in 13 major PK endpoints containing over 10,000 unique molecules. We built classification quantitative structure-activity relationship (QSAR) models for all but one endpoint (Cmax) following best practices of model development and validation. Those with acceptable external accuracy (CCR ≥ 0.60 and SE, PPV, SP, and NPV ≥ 0.50) include hepatic stability at 15, 30, and 60 minutes, hepatic half-life at the subcellular and tissue levels, renal clearance, blood brain barrier permeability, CNS activity, Caco-2 permeability, plasma protein binding, plasma half-life, microsomal intrinsic clearance, and oral bioavailability. As a case study to illustrate model utility, we employed all developed models to predict the PK properties of all compounds in DrugBank. We also predicted PK properties of molecules hitting popular drug targets among several organs SLC6A4 (brain), ADRB2 (heart and lungs), HMGCR (liver), and CaSR (kidneys) only. These analyses revealed that nearly all experimental, investigational, and withdrawn compounds included in DrugBank are hepatically stable at 60 minutes and under, exhibit CNS activity, and permeate the Caco-2 cell line (a measure of intestinal absorption). Furthermore, our results indicate that compounds targeting different organs have distinct predicted PK profiles. This observation suggests that desired PK properties depend on compound’s indication. All models described in this paper have been integrated and made publicly available via the novel predictor of pharmacokinetic properties (PhaKinPro) web portal that can be accessed at https://phakinpro.mml.unc.edu/.
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 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 | 0.001 | 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 | 0.001 | 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, 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".