Introduction to Regulatory Genomics and Epigenomics III: Hidden Markov Models in Gene Regulation

on YouTube ↗01h 38m

" summary

Covers using Hidden Markov Models (HMMs) to model gene regulation and epigenomic states, describing how HMMs are applied to genomic and epigenomic data (e.g., chromatin state annotation) and the algorithms for training and inference. Focuses on practical use of HMMs for regulatory genomics tasks such as segmenting genomes into functional states and interpreting regulatory patterns from sequencing-based assays.

AI-written summary · about

" chapters (33)

1. 0:00Intro
2. 0:35Hidden Markov Models
3. 4:03Outline
4. 5:33Markov Chains To compute the probability of a sequences of (arbitrary) length L, we can decompose it via the product rule
5. 7:48Training a Markov Chain A popular objective function to set parameters of Markov models is to maximize the likelihood of the data, defined as
6. 11:23Weight Matrices
7. 14:27Markov Chains as Stochastic Automata
8. 16:19Markov Chain: Formal Definition
9. 18:14Probability of a Sequence PYRYRYM
10. 18:58Another Example
11. 20:24Parsing: Finding the Most Probable Path
12. 21:43Parsing/Decoding with an HMM
13. 22:47Dynamic Programming: Best Partial Parse
14. 23:57The Viterbi Algorithm
15. 25:24Viterbi: Traceback
16. 26:20The Forward Algorithm : Probability of a Sequence
17. 30:32Example: Eukaryotic Gene Structure complete mRNA coding segment
18. 31:16Using an HMM for Gene Prediction
19. 32:26Higher Order Markovian Eukaryotic Recognizer (HOMER)
20. 36:41Unsupervised (ML) Training
21. 37:29Viterbi Training training features
22. 38:48Recall: The Forward Algorithm
23. 39:36The Backward Algorithm
24. 40:17Baum-Welch: Summing over All Paths
25. 42:49Combining Forward & Backward
26. 43:57Monotonic Convergence Behavior
27. 45:33Variants of Inference
28. 47:42Example: The Ahab model
29. 55:22HMMs & Geometric Feature Lengths
30. 57:42Recall: Decoding with an HMM
31. 58:16Recall: Viterbi Decoding for HMMs
32. 58:28Naive GHMM Decoding
33. 59:32Typical GHMM Topology for Gene Finding Fixed-length states are called signal

" related talks

• 38mInterpreting Deep Learning Models Of Functional Genomics Data To Decode Regulatory Sequence...Anshul Kundaje · Interpretable Machine Learning in Natural and Social Sciences (2022)
• 39mDynamics Programming for Machine Learning- Hidden Markov ModelsAvik Das · PyData Los Angeles 2019
• 35mTranscriptional Enhancers - Looking out for the Genes and Each OtherSridhar Hannenhalli · Network Biology 2016
• 01h 19mAnalyzing Genetic Networks Using Neural NetworksManu Flores · PyData Los Angeles 2019
• 43mMachine Learning in Molecular BiologyLeelavati Narlikar · Anthill Inside Miniconf – Pune 2017
• 01h 23mProbabilistic Graphical Models, HMMs using PGMPYHarish Kashyap K, Ria Aggarwal · ODSC India 2018
• 35mMachine Learning for Single-Cell 3D EpigenomicsJian Ma · From Algorithms to Discovery in Genome-Scale Biology and Medicine 2022
• 37mDynamics of 3D Enhancer-Gene Associations Across Diverse Human Cell Types and TissuesAnshul Kundaje · Network Biology 2016