
Robetta
A protein structure prediction service using deep learning and comparative modeling.

The Industry Standard for Protein Comparative Modeling and Structural Bio-Refinement.

MODELLER is a specialized computational tool used for homology or comparative modeling of protein three-dimensional structures. By 2026, it has solidified its position as a critical post-processing and refinement engine for AI-generated protein folds from systems like AlphaFold3 and RoseTTAFold. Unlike pure neural network predictors, MODELLER utilizes a technique known as 'satisfaction of spatial restraints'—it takes an alignment between a target sequence and known template structures as input and outputs a 3D model containing all non-hydrogen atoms. Technically, its architecture is built around a complex objective function that minimizes violations of restraints derived from the alignment and basic stereochemical rules. This makes it indispensable for researchers needing to include ligands, handle multi-component assemblies, or refine specific loops that general AI models may struggle with. Its Python-based scripting interface allows for high-level automation, making it a staple in high-throughput virtual screening and synthetic biology pipelines. While academics can access the tool for free, commercial licenses are strictly regulated, reflecting its high value in pharmaceutical R&D.
MODELLER is a specialized computational tool used for homology or comparative modeling of protein three-dimensional structures.
Explore all tools that specialize in model protein structures. This domain focus ensures MODELLER delivers optimized results for this specific requirement.
Explore all tools that specialize in homology modeling. This domain focus ensures MODELLER delivers optimized results for this specific requirement.
Uses a conjugate gradient and simulated annealing method to minimize an objective function based on probability density functions.
Discrete Optimized Protein Energy (DOPE) score for assessing the quality of atomic-level models based on a statistical potential.
Combines information from multiple structural templates to build a single consensus target model.
Automatic identification and refinement of flexible loop regions using ab initio methods within the framework.
Allows for the inclusion of HETATM records from templates into the target model.
Enforces non-crystallographic symmetry during the modeling of homomeric assemblies.
The core engine is accessible via a comprehensive Python API (Modeller module).
Obtain an academic or commercial license key from the Sali Lab website.
Install via Conda or binary package for Linux, Mac, or Windows.
Set the 'KEY_MODELLER' environment variable for authentication.
Identify homologous templates using tools like BLAST or HHpred.
Create a PIR-formatted alignment file between the target and templates.
Download corresponding PDB files for the templates from the RCSB PDB.
Write a Python script to define modeling parameters (e.g., number of models, refinement level).
Execute the script using the 'mod_python' or 'mod10.x' executable.
Analyze the output log for objective function values and DOPE scores.
Visualize the best-scoring models in PyMOL or ChimeraX for validation.
All Set
Ready to go
Verified feedback from other users.
"Highly reliable and technically robust, though the learning curve for the Python API is steep for non-programmers."
Post questions, share tips, and help other users.

A protein structure prediction service using deep learning and comparative modeling.

The Protein Engineering Operating System for targeting the undruggable proteome.

Gamified Protein Folding: Harnessing human intuition to solve complex biochemical puzzles for drug discovery.

The industry-standard engine for high-throughput genomic variant discovery and clinical-grade sequencing analysis.

Advanced gene function prediction and association network integration for precision genomics.

The premier open-source genomic workflow system for reproducible scientific research and advanced omics analysis.
The Human Protein Atlas is an open-access resource that maps all human proteins in cells, tissues, and organs.