Speaker
Description
The advent of high-throughput transcriptomics provides an opportunity to advance mechanistic understanding of transcriptional processes and their connections to cellular function at an unprecedented, genome-wide scale. These transcriptional systems, which involve discrete stochastic events, are naturally modeled using chemical master equations (CMEs), which can be solved for probability distributions to fit biophysical rates that govern system dynamics. While CME models have been used as standards in fluorescence transcriptomics for decades to analyze single-species RNA distributions, there are often no closed-form solutions to CMEs that model multiple species, such as nascent and mature RNA transcript counts. This has prevented the application of standard likelihood-based statistical methods for analyzing high-throughput, multi-species transcriptomic datasets using biophysical models. We show how neural networks and statistical understanding of system distributions can produce accurate approximations to a steady-state bivariate distribution of the bursty model of transcription, bypasses intensive numerical solving techniques and reducing likelihood evaluation time by several orders of magnitude. This method can be incorporated into existing machine learning architectures to enable broad exploration of parameter space of transcriptional burst sizes, RNA splicing rates, and mRNA degradation rates from experimental transcriptomic data.
Bibliography
@article{gorin_spectral_2024,
title = {Spectral neural approximations for models of transcriptional dynamics},
volume = {In press},
copyright = {Creative Commons Attribution 4.0 International License (CC-BY)},
url = {https://www.sciencedirect.com/science/article/abs/pii/S000634952400314X},
doi = {10.1016/j.bpj.2024.04.034},
language = {en},
urldate = {2022-06-22},
journal = {Biophysical Journal},
author = {Gorin, Gennady and Carilli, Maria and Chari, Tara and Pachter, Lior},
month = may,
year = {2024}
}
@article{carilli_biophysical_2024, title = {Biophysical modeling with variational autoencoders for bimodal, single-cell {RNA} sequencing data}, copyright = {All rights reserved}, url = {https://www.nature.com/articles/s41592-024-02365-9}, doi = {10.1038/s41592-024-02365-9}, language = {en}, urldate = {2023-01-16}, journal = {Nature Methods}, author = {Carilli, Maria T. and Gorin, Gennady and Choi, Yongin and Chari, Tara and Pachter, Lior}, month = jul, year = {2024} }
