Hidden Markov Models

The Viterbi Algorithm

Task

• In R, implement the Viterbi algorithm.

• Input:

  • A DNAString of observations.
  • Hidden Markov Model i.e. one of the prepared models HMM1.RData or HMM2.RData.

• Output:

  • Matrix of calculated probabilities for all observations generated by all hidden states (|N| x number of observations).
  • A vector of hidden states, that generated the input observations.

HMM = (N, M, A, B, pi)

• N set of hidden states
• M set of emitted characters
• A transition probability matrix
• B emission probability matrix
• pi initial probability distribution vector

Download files from GitHub

Basic Git settings

• Configure the Git editor

  git config --global core.editor notepad

• Configure your name and email address

  git config --global user.name "Zuzana Nova"
  git config --global user.email [email protected]

• Check current settings

  git config --global --list

• Create a fork on your GitHub account. On the GitHub page of this repository find a Fork button in the upper right corner.

• Clone forked repository from your GitHub page to your computer:

  git clone <fork repository address>

• In a local repository, set new remote for a project repository:

  git remote add upstream https://github.com/mpa-prg/exercise_11.git

Send files to GitHub

Create a new commit and send new changes to your remote repository.

• Add file to a new commit.

  git add <file_name>

• Create a new commit, enter commit message, save the file and close it.

  git commit

• Send a new commit to your GitHub repository.

  git push origin main