Structural biology and drug discovery

Homology Modeling and Its Use in Predicting Protein Structures

Hetal Marble, Ph.D.

Head of Commercial, Immuto Scientific

Proteins, one of the most important components of living organisms, are crucial to various functions. A thorough understanding of protein structures is essential for creating novel therapeutic approaches because of the intimate relationship between the protein's three-dimensional structure and its activity. Predicting protein structures is an essential goal in molecular biology and biochemistry because it is challenging and time-consuming to determine a protein's structure using experimental approaches. Homology modeling is just one example of a cutting-edge computational method that has recently been useful for predicting protein structures.

Homology modeling is a computational approach that uses the known structures of homologous proteins to predict the structure of a novel protein. Proteins with comparable activities tend to have similar structures, a hypothesis supported by the evolutionary conservation principle, which underlies this approach. The first step in building a homology model is to use sequence alignment techniques to find templates that have protein structures that are the closest match. A model of the new protein is then generated using the template structures and refined using various computational methods.

Homology modeling does have some restrictions, though. Prediction reliability depends on the integrity of used template structures and the precision of used sequence alignment algorithms. The absence of appropriate templates prevents the approach from predicting protein structures with original folds. As a result, the success of homology modeling relies on how well the template structure and the target protein structure match up. When the target protein and the template are highly similar, the predictions are extremely accurate; otherwise, the accuracy drops. Predictions can be improved when augmenting protein folding models with empirical data  - such as higher-order structure assessment using Immuto’s platform. Immuto’s approach measures directly where water molecules surround a protein, which contributes significantly to the Gibbs’ free energy of the fold and can vastly improve accuracy, especially for proteins that are novel or don’t have a great homology template.

Homology modeling can be used for more than only protein structure prediction. This ability to forecast protein complex and interaction architecture is crucial for diseases with broken protein complexes and linkages in their molecular pathways. To better understand the molecular basis of diseases and devise novel treatment techniques, scientists have used homology modeling to predict the shapes of proteins in bacteria and viruses.

In molecular biology and biochemistry, homology modeling can be used in a wide variety of contexts. It's a fast and cheap replacement for experimental ways of estimating the structures of undiscovered proteins. It has also been shown to be beneficial in drug discovery for predicting the binding of small compounds to proteins, which in turn facilitates the creation of novel therapies, and for understanding the consequences of mutations on protein structure and function.

These caveats notwithstanding, homology modeling is still a powerful method for predicting protein structures, complexes, and interactions. As computing power and algorithms advance, it will become increasingly important in molecular biology and biochemistry.

AUTHOR BIO

Hetal Marble, Ph.D.

Head of Commercial, Immuto Scientific

Dr. Hetal Marble is Head of Commercial for Immuto Scientific, a boutique CRO focused on protein structural characterization and drug discovery. Dr. Marble has had varied and extensive experience leading life science businesses. She led diagnostic product management at Quanterix Corporation, where she launched Quanterix’s first diagnostic test for Alzheimer’s disease and directed product strategy and go-to-market for all clinical diagnostic and research services products. She also led companion diagnostics and biomarker development for the Center for Integrated Diagnostics at MGH where she grew the division’s in-house CRO into a multi-million dollar enterprise by translating novel technologies into clinical practice and leading precision biomarker strategies for clinical trials. Prior to her work at MGH, Hetal worked in business development, alliance management, and strategic product marketing for both Promega Corporation and Pall Biotech (a Danaher company), managing and growing burgeoning commercial territories into millions of dollars in revenue. Hetal is foundationally an experienced research scientist – she received her Ph.D. at Brown University, where her graduate work focused on development of biomarkers for lineage-specific stem cell differentiation - she is the holder of two U.S. patents emerging from that work. Hetal is active in the scientific community; she is the elected Chair of Diversity, Equity, and Inclusion for the Association for Molecular Pathology (AMP); she is a member of the 2022 Women in Bio (WiB) cohort of Emerging Executives; she served on the Pathology Informatics Collaborative Community (PIcc), a collaborative community of the US Food and Drug Administration; and she is a contributing author on precision medicine for the book Taking Care of You, a book on women’s health being published by the Mayo Clinic Press in October 2022.

Related articles

Comparison with other techniques

X-Ray Crystallography: Limitations, Drawbacks and a Better Alternative

Structural biology and drug discovery

Affinity Maturation of Monoclonal Antibodies with Radical Footprinting

Structural biology and drug discovery

Structure-activity relationship (SAR) analysis and its applications in drug discovery