RESUMEN
Machine learning is a valuable tool in the development of chemical technologies but its applications into supramolecular chemistry have been limited. Here, the utility of kernel-based support vector machine learning using density functional theory calculations as training data is evaluated when used to predict equilibrium binding coefficients of small molecules with cucurbit[7]uril (CB[7]). We find that utilising SVMs may confer some predictive ability. This algorithm was then used to predict the binding of drugs TAK-580 and selumetinib. The algorithm did predict strong binding for TAK-580 and poor binding for selumetinib, and these results were experimentally validated. It was discovered that the larger homologue cucurbit[8]uril (CB[8]) is partial to selumetinib, suggesting an opportunity for tunable release by introducing different concentrations of CB[7] or CB[8] into a hydrogel depot. We qualitatively demonstrated that these drugs may have utility in combination against gliomas. Finally, mass transfer simulations show CB[7] can independently tune the release of TAK-580 without affecting selumetinib. This work gives specific evidence that a machine learning approach to recognition of small molecules by macrocycles has merit and reinforces the view that machine learning may prove valuable in the development of drug delivery systems and supramolecular chemistry more broadly.
Asunto(s)
Bencimidazoles/química , Hidrocarburos Aromáticos con Puentes/química , Compuestos Heterocíclicos con 3 Anillos/química , Imidazoles/química , Teoría Funcional de la Densidad , Modelos Químicos , Máquina de Vectores de SoporteRESUMEN
Measurements of neural responses to identically repeated experimental events often exhibit large amounts of variability. This noise is distinct from signal, operationally defined as the average expected response across repeated trials for each given event. Accurately distinguishing signal from noise is important, as each is a target that is worthy of study (many believe noise reflects important aspects of brain function) and it is important not to confuse one for the other. Here, we introduce a principled modeling approach in which response measurements are explicitly modeled as the sum of samples from multivariate signal and noise distributions. In our proposed method-termed Generative Modeling of Signal and Noise (GSN)-the signal distribution is estimated by subtracting the estimated noise distribution from the estimated data distribution. We validate GSN using ground-truth simulations and demonstrate the application of GSN to empirical fMRI data. In doing so, we illustrate a simple consequence of GSN: by disentangling signal and noise components in neural responses, GSN denoises principal components analysis and improves estimates of dimensionality. We end by discussing other situations that may benefit from GSN's characterization of signal and noise, such as estimation of noise ceilings for computational models of neural activity. A code toolbox for GSN is provided with both MATLAB and Python implementations.