Computational pipeline for detecting Kunitz-type protease inhibitor domains (PF00014) using a Profile HMM. Developed during the Laboratory of Bioinformatics 1 course (MSc in Bioinformatics, University of Bologna), this project integrates structural alignment, sequence clustering, and model evaluation for accurate protein domain annotation.

In this project we developed an HMM model for the Kuntiz domain to potentially annotate new sequences.

This repository contains the pipeline and files used to create a structure-informed profile Hidden Markov Model for the identification of the BPTI/Kunitz domain in protein sequences.

Building structure-informed Profile HMMs for Kunitz/BPTI-type protease inhibitor domain detection, comparing sequence-based vs. structure-based approaches with 2-fold cross-validation.

Development of a structure-driven HMM for the Kunitz domain (PF00014), combining curated 3D alignments and robust statistical evaluation. Project created during the MSc in Bioinformatics at the University of Bologna for the Laboratory of Bioinformatics 1 course.