Periodic Reporting for period 1 - BURnOUT (Enabling efficient cell engineering leaving gene-expression BURden OUT for cell therapies and biopharmaceutical industry)
Okres sprawozdawczy: 2024-09-01 do 2026-02-28
Mammalian cell engineering has been transforming modern biomedicine, enabling major advances in cell-based therapies such as CAR-T cells and in the production of complex biopharmaceuticals, including monoclonal antibodies. However, despite these developments, the design–build–test cycle for engineered cells has remained slow, costly, and inefficient. A key bottleneck has been the limited intracellular resources available for gene expression: when multiple transgenes were introduced, they competed for transcriptional and translational machinery, often resulting in imbalanced protein production and reduced cellular performance. This challenge has been particularly critical in emerging applications requiring coordinated expression of multiple genes, such as multi-target immunotherapies and combinatorial biologics.
The BURnOUT project has been addressing this unmet need by developing an artificial intelligence (AI) and machine learning (ML)-driven platform for the automated optimization of paired gene sequences, enabling balanced and efficient co-expression in mammalian cells. Building on prior experimental evidence that codon-level sequence design could modulate expression balance, the project has aimed to infer the underlying principles governing resource allocation and translate them into a predictive computational tool. By integrating systematic data generation with advanced sequence modeling, BURnOUT has been moving beyond traditional single-gene optimization toward a systems-level approach to multi-gene engineering.
The impact of BURnOUT is both scientific and translational. Scientifically, it contributes to advancing the understanding of gene expression burden and resource competition in mammalian systems, reinforcing links with synthetic and systems biology. Technologically, it delivers a novel platform capable of accelerating the design of engineered cells, thereby reducing development timelines and improving efficiency in preclinical pipelines. From an innovation perspective, BURnOUT has opened a new avenue in multi-gene optimization, with potential applications across cell therapy, biopharmaceutical manufacturing, and AI-driven life sciences.
The BURnOUT project has been addressing this unmet need by developing an artificial intelligence (AI) and machine learning (ML)-driven platform for the automated optimization of paired gene sequences, enabling balanced and efficient co-expression in mammalian cells. Building on prior experimental evidence that codon-level sequence design could modulate expression balance, the project has aimed to infer the underlying principles governing resource allocation and translate them into a predictive computational tool. By integrating systematic data generation with advanced sequence modeling, BURnOUT has been moving beyond traditional single-gene optimization toward a systems-level approach to multi-gene engineering.
The impact of BURnOUT is both scientific and translational. Scientifically, it contributes to advancing the understanding of gene expression burden and resource competition in mammalian systems, reinforcing links with synthetic and systems biology. Technologically, it delivers a novel platform capable of accelerating the design of engineered cells, thereby reducing development timelines and improving efficiency in preclinical pipelines. From an innovation perspective, BURnOUT has opened a new avenue in multi-gene optimization, with potential applications across cell therapy, biopharmaceutical manufacturing, and AI-driven life sciences.
The project has carried out an integrated set of experimental and computational activities to develop and validate the BURnOUT platform. It has generated a systematic dataset of paired gene sequences and corresponding expression profiles in mammalian cells, enabling the identification of key parameters governing balanced co-expression. Building on these data, the project has developed and trained a machine learning model capable of predicting optimized gene sequence pairs, which has been implemented into a user-oriented platform. The approach has then been validated experimentally in both biopharmaceutical (CHO cells) and therapeutic (primary T cells) contexts, demonstrating the ability to improve coordinated gene expression. In parallel, the project has advanced intellectual property positioning, market analysis, and exploitation strategies to support translation toward industrial and clinical applications.
While existing approaches primarily focus on optimizing single-gene expression using heuristic codon usage metrics, BURnOUT has introduced a fundamentally new paradigm by addressing the coordinated optimization of multiple genes, explicitly accounting for intracellular resource competition. Moreover, the development of an AI-driven platform capable of generating optimized gene pairs represents a significant advance over existing tools, which do not support simultaneous multi-gene optimization. This shift toward systems-level, data-driven design establishes a novel framework for engineering complex cellular functions, with direct implications for next-generation cell therapies and biomanufacturing.