Loading...
Modelo multivariado oculto de Markov expandido para capturar dinámicas de enzimas durante smFRET de libre difusión
Alzate Vargas, Sebastian
Alzate Vargas, Sebastian
Citations
Altmetric:
Abstract
Las enzimas cumplen una función muy importante como catalizadores, acelerando las reacciones químicas. Por ello, son fundamentales en los procesos bioquímicos. Los experimentos de transferencia de energía de resonancia de Forster de molécula única (smFRET) brindan la capacidad de modelar la dinámica fluxional en conformaciones enzimáticas. El smFRET de difusión libre registra las veces que los canales de fluoróforo aceptor y donante detectan los fotones mientras la difusión se produce a través de un sustrato con un láser confocal. Este tipo de experimento smFRET no limita el comportamiento dinámico de la molécula como lo hace el experimento smFRET inmovilizado en la superficie, pero los fotones registrados durante el experimento a menudo provienen del trasfondo y otros aspectos no relacionados con el comportamiento fluxional de la enzima. Proponemos un modelo de Markov oculto multivariante aumentado (AMHMM) para modelar simultáneamente los recuentos de fotones de los canales aceptor y donante mientras se agrega un estado markoviano auxiliar que representa que la enzima no está frente al láser. Las estimaciones iniciales de los parámetros se ajustan para extraer información del comportamiento dinámico de la molécula. Usando datos simulados, mostramos cómo nuestro modelo es capaz de capturar la realidad del terreno. Finalmente, aplicamos nuestro AMHMM para modelar el comportamiento dinámico de una enzima Alb3.
Enzymes serve a very important function as catalysts, accelerating chemical reactions. For this reason they are fundamental in biochemical processes. Single molecule Forster Resonance Energy Transfer (smFRET) experiments provide the capability of modeling the fluxional dynamics in enzyme conformations. Freely diffusing smFRET record the times photons are detected by acceptor and donor fluorophore channels while diffusion occurs through a substrate with a confocal laser. This type of smFRET experiment does not limit the dynamic behavior of the molecule as the surface immobilized smFRET experiment does, but the recorded photons during the experiment often come from the background and other aspects not related to the fluxional behavior of the enzyme. We propose an Augmented Multivariate Hidden Markov Model (AMHMM) to simultaneously model acceptor and donor channel photon counts while adding an auxiliary Markovian state that represents the enzyme not being in front of the laser. Initial parameter estimates are adjusted to extract information of the dynamic behavior of the molecule. Using simulated data, we show how our model is able to capture ground truth. Finally, we apply our AMHMM to model the dynamic behavior of an Alb3 enzyme.
Enzymes serve a very important function as catalysts, accelerating chemical reactions. For this reason they are fundamental in biochemical processes. Single molecule Forster Resonance Energy Transfer (smFRET) experiments provide the capability of modeling the fluxional dynamics in enzyme conformations. Freely diffusing smFRET record the times photons are detected by acceptor and donor fluorophore channels while diffusion occurs through a substrate with a confocal laser. This type of smFRET experiment does not limit the dynamic behavior of the molecule as the surface immobilized smFRET experiment does, but the recorded photons during the experiment often come from the background and other aspects not related to the fluxional behavior of the enzyme. We propose an Augmented Multivariate Hidden Markov Model (AMHMM) to simultaneously model acceptor and donor channel photon counts while adding an auxiliary Markovian state that represents the enzyme not being in front of the laser. Initial parameter estimates are adjusted to extract information of the dynamic behavior of the molecule. Using simulated data, we show how our model is able to capture ground truth. Finally, we apply our AMHMM to model the dynamic behavior of an Alb3 enzyme.
Description
Date
2023-12-08
Journal Title
Journal ISSN
Volume Title
Publisher
Collections
Keywords
Enzimas, smFRET, Modelo de Markov oculto multivariado