Computational Techniques for Protein 3D Structure Prediction

Name: Computational Techniques for Protein 3D Structure Prediction
Rating: 4.6 (29 reviews)

Learn how to Predict Protein 3D Structure using Computational Tools for Drug Discovery and more.

Created byShahroz Rahman

Last updated 10/2025

English

What you'll learn

Understand the fundamental principles of protein structure prediction.
Gain hands-on experience with leading computational tools such as MODELLER, Swiss-Model, I-TASSER, and AlphaFold.
Learn how to select appropriate templates, perform sequence alignment, and refine protein models for accuracy.
Explore advanced modeling options, model quality assessment techniques, and confidence score estimation methods.
Discover the applications of protein structure prediction in drug discovery, molecular biology, and protein engineering.
Stay updated with emerging trends and technologies in the field of computational bioinformatics.
Apply your newfound knowledge and skills to real-world research projects.

Course content

7 sections • 29 lectures • 5h 21m total length

Introduction7:24
Learn how computational techniques predict protein 3D structures from amino acid sequences, from homology modeling and threading to deep learning tools like AlphaFold, with hands-on workflows and validation.
Fundamentals of Protein Structure6:04
Explains protein structure from amino acids to four levels, including alpha helices and beta sheets, and how hydrogen bonds, ionic interactions, van der Waals forces, and disulfide bridges shape function.
Importance of Protein Structure Prediction10:14
Predict protein structures to reveal function and enable structure-based drug design by identifying binding sites and guiding lead discovery. Apply insights to enzyme engineering, protein therapeutics, and disease mechanisms.
Challenges in Protein Structure Prediction9:59
Explore the inherent complexity of proteins, including conformational flexibility, large multi-domain structures, and intrinsically disordered regions. Leverage machine learning, integrative modeling, and high-performance computing to address experimental and computational challenges.
Overview of Computational Techniques16:30
Explore four core computational techniques for protein structure prediction—Modeller, Swiss-model, I-tasser, and AlphaFold—plus other tools, their strengths, and applications.

Homology Modeling11:20
Learn homology modeling, or comparative modeling, to predict protein three-dimensional structures from templates using sequence similarity; apply in drug design, enzyme engineering, and functional insights.
Introduction to MODELLER10:16
Explore Modeller, a command-line, scriptable tool for protein homology modelling, guiding input preparation, sequence-structure alignment, model generation, refinement, and assessment for structure based drug design.
MODELLER Installation and Setup7:13
Install MODELLER and Python on your computer, download the software for your operating system, set up dependencies, and obtain a license key for homology modeling.
Template Selection and Alignment29:27
Learn template selection and alignment for protein structure prediction with modeller, using BLAST against the PDB database to pick templates, evaluate identity and coverage, and prepare alignment files.
Model Building10:58
Build models for homology modeling using a prepared alignment and template in Modeler, generate five models, compare dope and pdf scores, and select the best for downstream analysis.
Evaluation and Assessment of MODELLER Predictions20:10
Evaluate MODELLER predictions using DOPE score and GA 341, select the model with favorable scores, and validate with MolProbity, Verify3D, and ProCheck including Ramachandran plots.

Introduction to Swiss-Model7:25
Explore swiss-model, a web-based tool for homology modeling to predict protein structures. Learn its workflow—from template search and alignment to model building and evaluation with gmk and cumin scores.
Modeling Via Swiss-Model5:23
Perform homology modeling with Swiss-model by submitting a sequence, searching templates in the PDB database, selecting templates by identity and method, and building multiple models for downstream evaluation.
Model Quality Assessment4:47
Assess protein structures from Swiss-model using Ramachandran plots, Molprobity, and template alignment to gauge quality, then refine outliers with Wyncote for reliable downstream analysis.

Threading and Ab Initio Modeling12:22
Explore threading and ab initio modeling for protein structure prediction, contrast them with homology modeling, and understand when to use each method, along with their strengths and limitations.
Introduction to I-TASSER3:23
Learn the I-TASSER framework, an iterative threading assembly refinement server that combines ab initio and threading methods to predict, assess, and refine protein structures to high quality, with online options.
Structure Prediction20:52
Learn ab initio protein structure prediction with Itasser, input sequences, apply restraints and templates, run the job, and interpret three-dimensional models, scores, binding sites, and functional annotations.

Deep Learning Techniques8:16
Explore how deep learning powers protein structure prediction, from sequence embeddings and evolutionary information to 3D structures, through AlphaFold and other tools using transformer attention.
Introduction to AlphaFold13:53
Explore how AlphaFold uses AI, deep learning, and molecular biophysics to predict protein 3d structures from amino acid sequences, including MSA and end-to-end learning, and use AlphaFold server and database.
Protein Structure Prediction Using AlphaFold14:54
Predict protein structures using the AlphaFold server by submitting a sequence, evaluating confidence and expected position error, and downloading multiple models for visualization in Chimera or PyMOL.

Phyre2 - Remote Homology Detection and Modeling11:48
Learn how Phyre2 performs remote homology detection and template-based modeling to predict protein 3D structures, using hidden Markov models, threading, and ab initio loop modeling.
RaptorX10:03
Explore RaptorX, a deep-learning based protein structure prediction server offering distance-based modeling, inter-residue distance and orientation distributions, and downloadable 3D models for local use.
Rosetta - Fragment Based Modeling and Structure Prediction19:01
Predict protein structures ab initio and enable versatile modeling with Rosetta. Use Monte Carlo optimization and energy-based, knowledge-based scoring for docking, design, and mutational design.
Additional Tools for Protein Structure Prediction3:46

Integrative Modeling Approaches6:27
Integrative modeling combines multiple experimental and computational data sources to generate accurate, complete protein structures, using Bayesian inference, Monte Carlo sampling, and ensemble models.
Protein-Protein Docking7:50
Learn to predict how two proteins form a complex through docking, exploring rigid and flexible approaches, sampling poses, scoring, refinement, and validation against experimental evidence.
Protein Design and Engineering8:47
Explore protein design and engineering by creating novel sequences that fold into desired structures, using rational and de novo approaches, computational tools like Rosetta and Foldit, and machine learning design.
Structural Bioinformatics Databases10:55
Explore structural bioinformatics databases as digital archives of 3d coordinates and metadata, including PDB, Emdb, Scope, Cath, UniProt links, and the AlphaFold protein structure database, enabling modeling, benchmarking, and validation.
Trends and Technologies11:59
Explore how structural bioinformatics shifts from static structures to dynamic ensembles, guided by AI, cryoelectron microscopy, integrative modeling, and cloud-enabled, open science collaboration for proteome-scale insights.

Requirements

Basic knowledge of bioinformatics and protein biology concepts.
Access to a computer with internet connectivity.
Familiarity with basic computational concepts in Bioinformatics(preferred but not required).
Willingness to learn and engage in hands-on exercises.

Description

Welcome to "Computational Techniques for Protein 3D Structure Prediction," a cutting-edge course designed to equip you with the skills and knowledge needed to excel in the field of bioinformatics. If you're fascinated by the proteins and eager to explore the power of computational techniques in predicting their 3D structures, this course is for you.

In this course, we'll talk about the protein structure prediction, where you'll learn how to leverage advanced computational tools to predict the 3D structure of proteins form protein sequences. Through a series of engaging lectures and hands-on exercises, you'll gain a deep understanding of the fundamental principles, methodologies, and applications of protein structure prediction.

We'll kick off our journey with an insightful introduction, laying the groundwork by exploring the importance and challenges of protein structure prediction. From there, we'll dive into the practical aspects, we got four key sections:

Protein Structure Prediction Via MODELLER: Discover the power of MODELLER, a versatile software tool for homology modeling. Learn how to select appropriate templates, predict the proteins structure, and refine models to achieve accurate predictions.
Protein Structure Prediction Via Swiss-Model: Learn about the capabilities of Swiss-Model, an automated modeling platform renowned for its efficiency and reliability. Explore its features for automated modeling, advanced options, and model quality assessment.
Protein Structure Prediction Via I-TASSER: Explore the innovative approaches of I-TASSER, blending threading and ab initio modeling techniques for comprehensive structure prediction. We'll be talking about, obviously the structure prediction Via I-TASSER, model refinement, confidence score estimation, and comparative analysis.
Protein Structure Prediction Via AlphaFold (Machine Learning): Experience the future of protein structure prediction with AlphaFold, a groundbreaking deep learning algorithm. Understand its training data, model architecture, accuracy evaluation, and future applications.

Throughout the course, you'll not only learn the theoretical foundations but also gain practical skills through hands-on exercises, case studies, and real-world examples. Whether you're a seasoned bioinformatics researcher or a curious beginner, this course will empower you to master the art of protein structure prediction and unlock new opportunities in drug discovery, molecular biology, and beyond.

Join us on this exciting journey and become a proficient practitioner in the dynamic field of computational bioinformatics.

Enroll now! and take the first step towards mastering protein structure prediction!

Who this course is for:

Students and professionals in the fields of bioinformatics, computational biology, and molecular biology.
Researchers and scientists interested in leveraging computational techniques for protein structure prediction.
Individuals seeking to enhance their skills in computational modeling and bioinformatics analysis.
Anyone passionate about unraveling the mysteries of protein structures and exploring the intersection of biology and computer science.

Computational Techniques for Protein 3D Structure Prediction

What you'll learn

Explore related topics

Course content

Introduction5 lectures • 50min

Protein Structure Prediction Via MODELLER6 lectures • 1hr 29min

Protein Structure Prediction Via Swiss-Model3 lectures • 18min

Protein Structure Prediction Via I-TASSER3 lectures • 37min

Protein Structure Prediction Via AlphaFold (Machine Learning)3 lectures • 37min

Additional Tools for Protein 3D Structure Prediction4 lectures • 45min

Advanced Topics in Protein Structure Prediction5 lectures • 46min

Requirements

Description

Who this course is for: