Search

Recent advances in machine learning (ML) have led to newer model architectures including transformers (large language models, LLMs) showing state of the art results in text generation and image analysis as well as few-shot learning (FSLC) models which offer predictive power with extremely small datasets. These new architectures may offer promise, yet the 'no-free lunch' theorem suggests that no single model algorithm can outperform at all possible tasks. Here, we explore the capabilities of classical (SVR), FSLC, and transformer models (MolBART) over a range of dataset tasks and show a 'goldilocks zone' for each model type, in which dataset size and feature distribution (i.e. dataset "diversity") determines the optimal algorithm strategy. When datasets are small ( < 50 molecules), FSLC tend to outperform both classical ML and transformers. When datasets are small-to-medium sized (50-240 molecules) and diverse, transformers outperform both classical models and few-shot learning. Finally, when datasets are of larger and of sufficient size, classical models then perform the best, suggesting that the optimal model to choose likely depends on the dataset available, its size and diversity. These findings may help to answer the perennial question of which ML algorithm is to be used when faced with a new dataset., (© 2024. The Author(s).)

Butyrylcholinesterase (BChE) is a target of interest in late-stage Alzheimer's Disease (AD) where selective BChE inhibitors (BIs) may offer symptomatic treatment without the harsh side effects of acetylcholinesterase (AChE) inhibitors. In this study, we explore multiple machine learning strategies to identify BIs in silico , optimizing for precision over all other metrics. We compare state-of-the-art supervised contrastive learning (CL) with deep learning (DL) and Random Forest (RF) machine learning, across single and sequential modeling configurations, to identify the best models for BChE selectivity. We used these models to virtually screen a vendor library of 5 million compounds for BIs and tested 20 of these compounds in vitro . Seven of the 20 compounds displayed selectivity for BChE over AChE, reflecting a hit rate of 35% for our model predictions, suggesting a highly efficient strategy for modeling selective inhibition.

CYP2C19 is an important enzyme for organophosphate pesticide (OPP) metabolism. Because the OPPs can be both substrates and inhibitors of CYP2C19, we screened 45 OPPs for their ability to inhibit the activity of this enzyme and investigated the role of CYP2C19 in the metabolism of 22 of these molecules. We identified several nanomolar inhibitors of CYP2C19 as well as determined that thions, in general, are more potent inhibitors than oxons. We also determined that thions are readily metabolized by CYP2C19, although we saw no relationship between IC 50 values and intrinsic clearance rates. This study may have implications for mitigating the risk of OPP poisoning.

Pyronaridine, tilorone and quinacrine are cationic molecules that have in vitro activity against Ebola, SARS-CoV-2 and other viruses. All three molecules have also demonstrated in vivo activity against Ebola in mice, while pyronaridine showed in vivo efficacy against SARS-CoV-2 in mice. We have recently tested these molecules and other antivirals against human organic cation transporters (OCTs) and apical multidrug and toxin extruders (MATEs). Quinacrine was found to be an inhibitor of OCT2, while tilorone and pyronaridine were less potent, and these displayed variability depending on the substrate used. To assess whether any of these three molecules have other potential interactions with additional transporters, we have now screened them at 10 μM against various human efflux and uptake transporters including P-gp, OATP1B3, OAT1, OAT3, MRP1, MRP2, MRP3, BCRP, as well as confirmational testing against OCT1, OCT2, MATE1 and MATE2K. Interestingly, in this study tilorone appears to be a more potent inhibitor of OCT1 and OCT2 than pyronaridine or quinacrine. However, both pyronaridine and quinacrine appear to be more potent inhibitors of MATE1 and MATE2K. None of the three compounds inhibited MRP1, MRP2, MRP3, OAT1, OAT3, P-gp or OATP1B3. Similarly, we previously showed that tilorone and pyronaridine do not inhibit OATP1B1 and have confirmed that quinacrine behaves similarly. In total, these observations suggest that the three compounds only appear to interact with OCTs and MATEs to differing extents, suggesting they may be involved in fewer clinically relevant drug-transporter interactions involving pharmaceutical substrates of the other major transporters tested., Competing Interests: The authors declare the following competing financial interest(s): S.E. is founder and owner of Collaborations Pharmaceuticals, Inc. P.A.V., T.R.L. and A.C.P. are employees. The company has patents filed for the antiviral and other activities of these compounds. All other authors have no conflicts of interest., (© 2023 The Authors. Published by American Chemical Society.)

Acetylcholinesterase (AChE) is an important enzyme and target for human therapeutics, environmental safety, and global food supply. Inhibitors of this enzyme are also used for pest elimination and can be misused for suicide or chemical warfare. Adverse effects of AChE pesticides on nontarget organisms, such as fish, amphibians, and humans, have also occurred as a result of biomagnifications of these toxic compounds. We have exhaustively curated the public data for AChE inhibition data and developed machine learning classification models for seven different species. Each set of models were built using up to nine different algorithms for each species and Morgan fingerprints (ECFP6) with an activity cutoff of 1 μM. The human (4075 compounds) and eel (5459 compounds) consensus models predicted AChE inhibition activity using external test sets from literature data with 81% and 82% accuracy, respectively, while the reciprocal cross (76% and 82% percent accuracy) was not species-specific. In addition, we also created machine learning regression models for human and eel AChE inhibition to return a predicted IC 50 value for a queried molecule. We did observe an improved species specificity in the regression models, where a human support vector regression model of human AChE inhibition (3652 compounds) predicted the IC 50 s of the human test set to a better extent than the eel regression model (4930 compounds) on the same test set, based on mean absolute percentage error (MAPE = 9.73% vs 13.4%). The predictive power of these models certainly benefits from increasing the chemical diversity of the training set, as evidenced by expanding our human classification model by incorporating data from the Tox21 library of compounds. Of the 10 compounds we tested that were predicted active by this expanded model, two showed >80% inhibition at 100 μM. This machine learning approach therefore offers the ability to rapidly score massive libraries of molecules against the models for AChE inhibition that can then be selected for future in vitro testing to identify potential toxins. It also enabled us to create a public website, MegaAChE, for single-molecule predictions of AChE inhibition using these models at megaache.collaborationspharma.com.

Acetylcholinesterase (AChE) is an important drug target in neurological disorders like Alzheimer's disease, Lewy body dementia, and Parkinson's disease dementia as well as for other conditions like myasthenia gravis and anticholinergic poisoning. In this study, we have used a combination of high-throughput screening, machine learning, and docking to identify new inhibitors of this enzyme. Bayesian machine learning models were generated with literature data from ChEMBL for eel and human AChE inhibitors as well as butyrylcholinesterase inhibitors (BuChE) and compared with other machine learning methods. High-throughput screens for the eel AChE inhibitor model identified several molecules including tilorone, an antiviral drug that is well-established outside of the United States, as a newly identified nanomolar AChE inhibitor. We have described how tilorone inhibits both eel and human AChE with IC 50 's of 14.4 nM and 64.4 nM, respectively, but does not inhibit the closely related BuChE IC 50 > 50 μM. We have docked tilorone into the human AChE crystal structure and shown that this selectivity is likely due to the reliance on a specific interaction with a hydrophobic residue in the peripheral anionic site of AChE that is absent in BuChE. We also conducted a pharmacological safety profile (SafetyScreen44) and kinase selectivity screen (SelectScreen) that showed tilorone (1 μM) only inhibited AChE out of 44 toxicology target proteins evaluated and did not appreciably inhibit any of the 485 kinases tested. This study suggests there may be a potential role for repurposing tilorone or its derivatives in conditions that benefit from AChE inhibition.

Tuberculosis is caused by Mycobacterium tuberculosis (Mtb) and is a deadly disease resulting in the deaths of approximately 1.5 million people with 10 million infections reported in 2018. Recently, a key condensation step in the synthesis of mycolic acids was shown to require β-ketoacyl-ACP synthase (KasA). A crystal structure of KasA with the small molecule DG167 was recently described, which provided a starting point for using computational structure-based approaches to identify additional molecules binding to this protein. We now describe structure-based pharmacophores, docking and machine learning studies with Assay Central as a computational tool for the identification of small molecules targeting KasA. We then tested these compounds using nanoscale differential scanning fluorimetry and microscale thermophoresis. Of note, we identified several molecules including the Food and Drug Administration (FDA)-approved drugs sildenafil and flubendazole with K d values between 30-40 μM. This may provide additional starting points for further optimization., Competing Interests: The authors declare the following competing financial interest(s): S.E. is the CEO of Collaborations Pharmaceuticals, Inc. A.C.P., T.R.L., P.A.V., and K.M.Z. are employees at Collaborations Pharmaceuticals, Inc. G.C.C., M.B.N., and J.S.F. do not have conflicts of interest., (© 2020 American Chemical Society.)

Alzheimer's disease (AD) is the most common cause of dementia, affecting approximately 35 million people worldwide. The current treatment options for people with AD consist of drugs designed to slow the rate of decline in memory and cognition, but these treatments are not curative, and patients eventually suffer complete cognitive injury. With the substantial amounts of published data on targets for this disease, we proposed that machine learning software could be used to find novel small-molecule treatments that can supplement the AD drugs currently on the market. In order to do this, we used publicly available data in ChEMBL to build and validate Bayesian machine learning models for AD target proteins. The first AD target that we have addressed with this method is the serine-threonine kinase glycogen synthase kinase 3 beta (GSK3β), which is a proline-directed serine-threonine kinase that phosphorylates the microtubule-stabilizing protein tau. This phosphorylation prompts tau to dissociate from the microtubule and form insoluble oligomers called paired helical filaments, which are one of the components of the neurofibrillary tangles found in AD brains. Using our Bayesian machine learning model for GSK3β consisting of 2368 molecules, this model produced a five-fold cross validation ROC of 0.905. This model was also used for virtual screening of large libraries of FDA-approved drugs and clinical candidates. Subsequent testing of selected compounds revealed a selective small-molecule inhibitor, ruboxistaurin, with activity against GSK3β (avg IC 50 = 97.3 nM) and GSK3α (IC 50 = 695.9 nM). Several other structurally diverse inhibitors were also identified. We are now applying this machine learning approach to additional AD targets to identify approved drugs or clinical trial candidates that can be repurposed as AD therapeutics. This represents a viable approach to accelerate drug discovery and do so at a fraction of the cost of traditional high throughput screening., Competing Interests: The authors declare the following competing financial interest(s): SE is founder and owner of Collaborations Pharmaceuticals, Inc. PAV, ACP, DHF and EM are employees of Collaborations Pharmaceuticals, Inc., (© 2020 American Chemical Society.)

Durable gene silencing through the formation of compact heterochromatin domains plays a critical role during mammalian development in establishing defined tissues capable of retaining cellular identity. Hallmarks of heterochromatin gene repression are the binding of heterochromatin protein 1 (HP1), trimethylation of lysine 9 on histone H3 (H3K9me3) and the methylation of cytosine residues of DNA. HP1 binds directly to the H3K9me3 histone modification, and while DNA methyltransferases have been found in complex with histone methyltransferases and HP1, there remains much to be known about the relationship between DNA sequence and HP1 in differentiated mammalian cells. To further explore this interplay in a controlled system, we designed a system to test the effect of promoter CpG content on the formation kinetics and memory of an HP1-mediated heterochromatin domain in mouse embryo fibroblasts (MEF)s. To do this, we have constructed a side-by-side comparison of wild-type (CpGFull) and CpG-depleted (CpGDep) promoter-driven reporter constructs in the context of the Chromatin in vivo Assay (CiA), which uses chemically-induced proximity (CIP) to tether the chromoshadow domain of HP1α (csHP1α) to a fluorescent reporter gene in a reversible, chemically-dependent manner. By comparing the response of CpGFull and CpGDep reporter constructs, we discovered that the heterochromatin formation by recruitment of csHP1α is unaffected by the underlying CpG dinucleotide content of the promoter, as measured by the velocity of gene silencing or enrichment of H3K9me3 at the silenced gene. However, recovery from long-term silencing is measurably faster in the CpG-depleted reporter lines. These data provide evidence that the stability of the HP1 heterochromatin domain is reliant on the underlying DNA sequence. Moreover, these cell lines represent a new modular system with which to study the effect of the underlying DNA sequences on the efficacy of epigenetic modifiers., Competing Interests: The authors have declared that no competing interests exist.