jpHMM

jpHMM applies a jumping profile Hidden Markov Model to detect recombination breakpoints and assign subtypes in HIV-1 and HBV genomic sequences.

Key Features:

• Jumping profile HMM approach: Extends traditional alignment methods by incorporating both vertical and horizontal information from multiple sequence alignments.
• Subtyping and phylogenetic breakpoint prediction: Assigns regions of input sequences to known HIV-1 or HBV subtypes using a pre-calculated multiple alignment and predicts recombination breakpoints.
• Enhanced recombination detection: Accommodates circular genome structures and models dependencies between nucleotides at the 5'- and 3'-ends to improve breakpoint detection.
• Uncertainty assessment: Computes posterior probabilities and reports uncertainty regions and interval estimates for predicted breakpoints.
• Circular genome analysis extension: Includes a circular-genome version that avoids bias against breakpoints near sequence termini and does not require a fixed origin for coordinates.

Scientific Applications:

• Epidemiological studies: Enables detection and tracking of recombinant HIV-1 and HBV strains for epidemiological investigations.
• Vaccine development: Provides recombination and subtype information relevant to antigen selection and diversity assessment for vaccine research.
• Viral evolution and genetic diversity: Supports analyses of viral evolution and genetic diversity by identifying recombination events and subtype structure.
• Subtype classification and recombination mapping: Facilitates subtype assignment and precise localization of phylogenetic breakpoints in HIV-1 and HBV genomes.

Methodology:

Implements a jumping profile Hidden Markov Model using pre-calculated multiple alignments of major HIV-1 or HBV subtypes, computes posterior probabilities to derive breakpoint interval estimates and uncertainty regions, and includes an extension for circular genomes that models 5'-3' nucleotide dependencies.