|
Activity Number:
|
34
- Advanced Methods in Statistical Learning
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 7, 2022 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Learning and Data Science
|
|
Abstract #323338
|
|
|
Title:
|
An Augmented Multivariate Hidden Markov Model to Capture Dynamics in Freely Diffusing SmFRET Experiments
|
|
Author(s):
|
Axel Cortes and Roberto Rivera* and Sebastian Alzate
|
|
Companies:
|
University of Puerto Rico-Mayaguez and University of Puerto Rico-Mayaguez and University of Puerto Rico-Mayaguez
|
|
Keywords:
|
freely diffusing smFRET;
Markovian;
auxiliary state;
enzyme;
catalysis;
HMM
|
|
Abstract:
|
Proteins play a fundamental role in catalyzing reactions in living organisms. Without enzymes, i.e.,proteins with catalytic function, most biochemical reactions will not occur under physiological conditions. Single-molecule Forster resonance energy transfer (smFRET) experiments gives us the ability to model dynamic changes in enzyme configurations. Freely-diffusing smFRET methods record photon emission timestamps from labeled molecules as they diffuse through a solution with a confocal laser spot focused inside the solution. This version of smFRET does not limit the dynamic behavior of the molecule like the surface immobilized version does, but a side effect is that the recorded data frequently does not come from the molecule. We propose an Augumented Multivariate Hidden Markov Model (AMHMM); with an auxiliary state that represents when the data comes strictly from the background. Initial parameter estimates are later adjusted to extract insight about the molecule's fluxional behavior. Using simulated data we show how our model succeeds in capturing fluxional ground truth. This is followed by applying our model to real experimental data.
|
Authors who are presenting talks have a * after their name.
