7: Hidden Markov Models I
( \newcommand{\kernel}{\mathrm{null}\,}\)
Hidden Markov Models (HMMs) are a fundamental tool from machine learning that is widely used in computational biology. Using HMMs, we can explore the underlying structure of DNA or polypeptide sequences, detecting regions of especial interest. For instance, we can identify conserved subsequences or uncover regions with different distributions of nucleotides or amino acids such as promoter regions and CpG islands. Using this probabilistic model, we can illuminate the properties and structural components of sequences and locate genes and other functional elements.