Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
Deep generative models can generate synthetic data to tackle challenges inherent in real-world data within bioengineering and medicine. These challenges include concerns around privacy, biases in data, distributional shifts, underrepresentation of specific populations, and the scarcity of high-quality data. See Boris van Breugel et al
Biased and unrepresentative scientific data can lead to misleading conclusions and potentially harm patients. Artificial intelligence (AI) might be able to help make data more representative, but only if a standardized approach to assessing the quality of AI-generated data is established.
The FDA Modernization Act 2.0 allows bioengineered models in drug testing, including benchtop testing and computational models. These alternatives to animal testing may particularly advance drug discovery in reproductive and pregnancy research.
Stress urinary incontinence (SUI), a frequently underdiagnosed condition that mainly affects women, lacks effective and long-term treatment options. MUVON Therapeutics has developed a tissue-engineered advanced therapy medicinal product for the treatment of SUI, based on autologous cells, which is being tested in a phase II clinical study — a challenging development effort.
Synthetic data can be created by deep generative models to address challenges associated with real data, such as privacy issues, bias and data scarcity. This Review discusses the generation and application of synthetic data in biomedicine and bioengineering, including quality assessment and validation.
Synthetic gene circuits can endow cells with therapeutic functions. This Review discusses synthetic macromolecular systems, encompassing transcriptional, translational and post-translational mechanisms, that conditionally regulate protein expression and activity in response to specific internal or external stimuli, and their role in enhancing the therapeutic efficacy and safety of engineered cell therapies.
Volumetric compression is a pervasive phenomenon in the human body, manifesting during development, limb movement, digestion, tumorigenesis and injury. This Review provides an in-depth discussion of emerging engineering methods centred on volumetric compression, including foundational rationales, design principles and illustrative applications.
Microbial catalysts must partition incoming substrate between synthesis of biomass and synthesis of a desired product. Two-stage bioprocesses can accommodate this tradeoff to maximize process productivity by temporal separation of growth and production phases. This Review discusses the challenges of maintaining a high metabolic activity during the production phase.