Diversity
Explore a broad repertoire of phage-derived proteins, including endolysins, depolymerases and virion-associated lysins.
ANID Exploración 2024 · Project 13240071
Synthetic phage-based tools
Exploring phage-derived lytic systems for bacterial pathogen control
Integrating biological data, machine learning, genome mining and experimental validation to explore and prioritize phage-derived antibacterial candidates.
Phage biology
Protein engineering
Machine learning
Aquaculture pathogens
PhageLys
From biological diversity to experimentally testable candidates.
The challenge
Finding promising lytic proteins requires more than sequence similarity.
Phage-derived proteins occupy a large and heterogeneous functional space. PhageLys combines evidence-aware data, computational prioritization and experimental work to make candidate selection more systematic and traceable.
Evidence
Keep functional annotations and evidence levels visible when integrating and comparing candidate proteins.
Translation
Use computation to reduce the search space while keeping experimental validation as the test of biological activity.
Project architecture
One discovery ecosystem, multiple connected layers.
The project connects genome mining and curated data with functional classification, multidimensional characterization, candidate discovery and experimental validation.
The PhageLys ecosystem
Five components, each focused on a distinct part of the search problem.
Each component can evolve independently while remaining part of a shared discovery workflow.
PhageLysData
Evidence-aware resource integrating phage-derived proteins, functional annotations, structural information and biological evidence.
PhageLysClass
Machine-learning system for recognizing phage lytic proteins and assigning major functional classes.
PhageLysCharacterization
Multidimensional characterization of candidates using sequence, physicochemical, structural and representation-based information.
PhageLysDiscovery
Exploration and prioritization workflows for comparing candidates with evidence-supported proteins and identifying promising regions of protein space.
BactoMobilome
Reproducible workflows for recovering phage-associated genomic regions and proteins from bacterial genomes and comparing candidates across cohorts.
Discovery workflow
From biological information to focused experimental testing.
Predictions support prioritization. They do not replace experimental evidence of antibacterial activity.
Explore
Integrate phage knowledge and bacterial genomes.
Detect
Recognize lytic proteins and functional classes.
Characterize
Compare sequence, properties and structure.
Prioritize
Reduce the search space to focused candidates.
Test
Express, purify and experimentally evaluate.
759K+unique proteins
7integrated data sources
11K+evidence-supported proteins
5ecosystem components
Research & resources
Open resources supporting the project.
Repository
PhageLysData
Data construction workflow and project resources.
↗ bioRxiv · 2026PhageLysData preprint
Evidence-aware and AI-ready dataset of phage lytic enzymes and depolymerases.
↗
In development
PhageLys software ecosystem
Classification, characterization, discovery and bacterial mobilome mining components.
•••Project leadership
A multidisciplinary effort connecting phage biology, biotechnology and computational discovery.
The project combines experimental and computational research around synthetic phage-based approaches for bacterial pathogen control.
Dr. Jaime Romero
jromero@inta.uchile.clDr. María Elena Lienqueo
mlienque@ing.uchile.clANID
Funded project
ANID Exploración 2024 · Project 13240071
EXPLORING SYNTHETIC PHAGE-BASED TOOLS FOR CONTROLLING BACTERIAL PATHOGENS