Scientists at UCLA and the University of Toronto have unveiled a groundbreaking computational tool named moPepGen, designed to revolutionize how researchers identify genetic mutations at the protein level. This innovation, detailed in a recent publication in Nature Biotechnology, promises to unravel previously invisible variations in proteins, offering new insights into cancer biology and other complex diseases. By addressing an enduring bottleneck in proteogenomics, moPepGen stands to transform our understanding of how DNA alterations translate into functional protein changes that drive disease progression.
Proteogenomics, the interdisciplinary field combining genomics and proteomics, provides a comprehensive molecular portrait by linking genetic information to the proteome — the vast array of proteins expressed in cells. However, existing analytical tools have struggled to accurately detect variant peptides that arise from genetic mutations, alternative splicing events, and other sophisticated modifications. This limitation has curtailed efforts to map how mutations manifest at the protein level, leaving critical disease-associated changes undetectable in standard workflows. moPepGen overcomes this challenge by enabling precise detection of a broad spectrum of protein variants, thereby opening new avenues for diagnostic and therapeutic discovery.
The fundamental difficulty moPepGen addresses lies in the extraordinary complexity of genetic and transcriptomic variations that influence protein sequences. Conventional methods mostly detect simple amino acid substitutions, missing a plethora of protein forms generated by mechanisms such as alternative splicing, circular RNA translation, RNA editing, and gene fusions. These complex modifications have repeatedly been shown to contribute significantly to disease phenotypes but have remained largely hidden due to analytical constraints. moPepGen’s innovative design incorporates a graph-based computational model that systematically captures and interprets these diverse genetic alterations, providing a panoramic view of proteomic diversity.
Developed through an interdisciplinary collaboration, moPepGen employs an efficient algorithm capable of rapidly processing massive datasets derived from multiple technologies and species. According to Chenghao Zhu, PhD, co-first author of the study and a postdoctoral scholar at UCLA’s department of human genetics, this tool allows researchers to discern which genetic variants are genuinely expressed at the protein level—a capability that has been elusive until now. The algorithm’s speed and versatility enable it to handle the immense volumes of data generated by modern proteogenomic experiments while maintaining a high resolution of variant detection.
In demonstration of its capabilities, the research team applied moPepGen to proteogenomic datasets derived from a diverse cohort including prostate and kidney tumor samples along with hundreds of cancer cell lines. This rigorous testing confirmed moPepGen’s superior sensitivity and comprehensiveness, identifying four times more unique protein variants than prior methodologies. These newly detected variants encompassed a wide array of disease-relevant modifications, including those resulting from gene fusions and RNA editing events, which had previously evaded detection. This enhanced discovery pipeline not only augments the depth of molecular profiling but also refines our understanding of tumor heterogeneity and disease mechanisms.
One of the most promising applications of moPepGen lies in the burgeoning field of cancer immunotherapy. The tool can identify tumor-specific variant peptides that serve as neoantigens — unique markers not found in normal cells that are essential for designing personalized cancer vaccines and adoptive cell therapies. By systematically cataloging these neoantigens, moPepGen facilitates the development of targeted immunotherapies tailored to the unique proteomic landscape of an individual’s tumor, potentially improving clinical outcomes and minimizing off-target effects. This capability heralds a new era of precision oncology where treatment is directly informed by the intricate molecular signatures of cancer.
Beyond oncology, moPepGen offers transformative potential for studying neurodegenerative diseases and other conditions where protein alterations drive pathology. The ability to detect previously invisible variants enhances the resolution of disease-associated protein changes, illuminating novel mechanisms that could be therapeutically exploited. Its open-access availability ensures that researchers worldwide can integrate moPepGen into existing proteomic workflows, democratizing access to state-of-the-art computational analyses and accelerating discovery across multiple biomedical disciplines.
The tool’s underlying graph-based approach models gene expression and translation with unparalleled sophistication. Unlike traditional linear reference databases, moPepGen constructs a network capturing all possible variant sequences encoded by complex genetic events. This comprehensive mapping allows it to trace the translation of diverse transcript isoforms into their protein products, a feat that significantly improves variant identification accuracy. Such granular understanding of protein variantomes is critical as proteins are the primary effectors of cellular function and represent the direct interface where genetic mutations exert phenotypic effects.
Notably, moPepGen’s compatibility with multiple organismal genomes and proteomic technologies highlights its versatility and broad applicability. Whether analyzing human tumor samples or model organisms, researchers can leverage the tool’s robust platform to gain insights into protein variation landscapes under diverse biological contexts. Its capacity to scale with large datasets also aligns well with the current trajectory of big-data biology, where high-throughput sequencing and mass spectrometry produce massive volumes of complex data requiring sophisticated computational handling.
The development of moPepGen represents a significant step forward in overcoming challenges that have impeded proteogenomic research. Proteins operate as pivotal mediators of biological function, and subtle alterations in their sequences can have profound impacts on cellular behavior and disease progression. By illuminating these subtle yet critical protein variants, moPepGen bridges a crucial gap between genomic data and functional protein expression. This linkage not only deepens our biological understanding but also enhances the precision of molecular diagnostics and therapeutics.
The collaborative nature of this research, combining expertise from UCLA and the University of Toronto, underscores the interdisciplinary and international efforts driving proteogenomic innovation. Co-senior authors Paul Boutros, PhD, and Thomas Kislinger, PhD, have shepherded this project to fruition, emphasizing an integrative scientific approach that couples computational innovation with clinical relevance. Their work exemplifies how advanced bioinformatics tools can transform raw molecular data into actionable biomedical insights.
Researchers interested in employing moPepGen can access the tool freely on GitHub, where it integrates seamlessly with existing proteomics pipelines. This openness promotes widespread adoption and encourages continuous development within the scientific community. The availability of such a sophisticated, yet user-friendly, resource fosters an environment where cutting-edge proteogenomic analysis becomes standard practice, accelerating the translation of molecular discoveries into tangible health benefits.
As precision medicine continues to evolve, tools like moPepGen will be indispensable in decoding the molecular intricacies underpinning complex diseases. By capturing the full spectrum of protein variations that stem from genetic mutations and transcriptomic alterations, this tool enhances our capacity to identify novel biomarkers and drug targets. Its ability to reveal protein diversity in unprecedented detail paves the way for more accurate diagnostics, personalized therapies, and ultimately, improved patient outcomes across oncology and beyond.
Subject of Research: Proteogenomics, protein variant detection, cancer genomics, computational biology
Article Title: moPepGen: A Graph-Based Computational Tool for Comprehensive Detection of Protein Variants in Proteogenomics
News Publication Date: 2025
Web References:
- moPepGen GitHub: https://github.com/uclahs-cds/package-moPepGen
- Nature Biotechnology article: https://www.nature.com/articles/s41587-025-02701-0
References:
Zhu, C., Liu, L., Boutros, P., Kislinger, T., et al. (2025). moPepGen: A graph-based approach to discovering protein variants from complex genetic alterations. Nature Biotechnology. https://doi.org/10.1038/s41587-025-02701-0
Keywords: Cancer genomics, Genetics, Protein functions, Phenotypic variation
