Transfer learning Bayesian optimization for competitor DNA molecule design for use in diagnostic assays

With the rise in engineered biomolecular devices, there is an increased need for tailor-made biological sequences. Often, many similar biological sequences need to be made for a specific application meaning numerous, sometimes prohibitively expensive, lab experiments are necessary for their optimization. This paper presents a transfer learning design of experiments workflow to make this development feasible. By combining a transfer learning surrogate model with Bayesian optimization, we show how the total number of experiments can be reduced by sharing information between optimization tasks. We demonstrate the reduction in the number of experiments using data from the development of DNA competitors for use in an amplification-based diagnostic assay. We use cross-validation to compare the predictive accuracy of different transfer learning models, and then compare the performance of the models for both single objective and penalized optimization tasks.

Harnessing the power of machine learning into tissue engineering: current progress and future prospects

Tissue engineering is a discipline based on cell biology and materials science with the primary goal of rebuilding and regenerating lost and damaged tissues and organs. Tissue engineering has developed rapidly in recent years, while scaffolds, growth factors, and stem cells have been successfully used for the reconstruction of various tissues and organs. However, time-consuming production, high cost, and unpredictable tissue growth still need to be addressed. Machine learning is an emerging interdisciplinary discipline that combines computer science and powerful data sets, with great potential to accelerate scientific discovery and enhance clinical practice. The convergence of machine learning and tissue engineering, while in its infancy, promises transformative progress. This paper will review the latest progress in the application of machine learning to tissue engineering, summarize the latest applications in biomaterials design, scaffold fabrication, tissue regeneration, and organ transplantation, and discuss the challenges and future prospects of interdisciplinary collaboration, with a view to providing scientific references for researchers to make greater progress in tissue engineering and machine learning.

Mechanobiological optimization of scaffolds for bone tissue engineering

Synthetic bone graft scaffolds aim to generate new bone tissue and alleviate the limitations of autografts and allografts. To meet that aim, it is essential to have a design approach able to generate scaffold architectures that will promote bone formation. Here, we present a topology-varying design optimization method, the "mixed-topology" approach, that generates new designs from a set of starting structures. This approach was used with objective functions focusing on improving the scaffold's local mechanical microenvironments to mechanobiologically promote bone formation within the scaffold and constraints to ensure manufacturability and achieve desired macroscale properties. The results demonstrate that this approach can successfully generate scaffold designs with improved microenvironments, taking into account different combinations of relevant stimuli and constraints.

Sustainable biofabrication: from bioprinting to AI-driven predictive methods

Biofabrication is potentially an inherently sustainable manufacturing process of bio-hybrid systems based on biomaterials embedded with cell communities. These bio-hybrids promise to augment the sustainability of various human activities, ranging from tissue engineering and robotics to civil engineering and ecology. However, as routine biofabrication practices are laborious and energetically disadvantageous, our society must refine production and validation processes in biomanufacturing. This opinion highlights the research trends in sustainable material selection and biofabrication techniques. By modeling complex biosystems, the computational prediction will allow biofabrication to shift from an error-trial method to an efficient, target-optimized approach with minimized resource and energy consumption. We envision that implementing bionomic rationality in biofabrication will render bio-hybrid products fruitful for greening human activities.

Experimentally characterizing the spatially varying anisotropic mechanical property of cancellous bone via a Bayesian calibration method

Traditional experimental tests for characterizing bone's mechanical properties usually hypothesize a uniaxial stress condition without quantitatively evaluating the influence of spatially varying principal material orientations, which cannot accurately predict the mechanical properties distribution of bones in vivo environment. In this study, a Bayesian calibrating procedure was developed using quantified multiaxial stress to investigate cancellous bone's local anisotropic elastic performance around joints as the spatial variation of main bearing orientations. First, the bone cube specimens from the distal femur of sheep are prepared using traditional anatomical axes. The multiaxial stress state of each bone specimen is calibrated using the actual principal material orientations derived from fabric tensor at different anatomical locations. Based on the calibrated multiaxial stress state, the process of identifying mechanical properties is described as an inverse problem. Then, a Bayesian calibration procedure based on a surrogate constitutive model was developed via multiaxial stress correction to identify the anisotropic material parameters. Finally, a comparison between the experiment and simulation results is discussed by applying the optimal model parameters obtained from the Bayesian probability distribution. Compared to traditional uniaxial methods, our results prove that the calibration based on the spatial variation of the main bearing orientations can significantly improve the accuracy of characterizing regional anisotropic mechanical responses. Moreover, we determine that the actual mechanical property distribution is influenced by complicated mechanical stimulation. This study provides a novel method to evaluate the spatially varying mechanical properties of bone tissues enduring complex mechanical loading accurately and effectively. It is expected to provide more realistic mechanical design targets in vivo for a personalized artificial bone prosthesis in clinical treatment.

Personalized Volumetric Tissue Generation by Enhancing Multiscale Mass Transport through 3D Printed Scaffolds in Perfused Bioreactors

Engineered tissues provide an alternative to graft material, circumventing the use of donor tissue such as autografts or allografts and non-physiological synthetic implants. However, their lack of vasculature limits the growth of volumetric tissue more than several millimeters thick which limits their success post-implantation. Perfused bioreactors enhance nutrient mass transport inside lab-grown tissue but remain poorly customizable to support the culture of personalized implants. Here, a multiscale framework of computational fluid dynamics (CFD), additive manufacturing, and a perfusion bioreactor system are presented to engineer personalized volumetric tissue in the laboratory. First, microscale 3D printed scaffold pore geometries are designed and 3D printed to characterize media perfusion through CFD and experimental fluid testing rigs. Then, perfusion bioreactors are custom-designed to combine 3D printed scaffolds with flow-focusing inserts in patient-specific shapes as simulated using macroscale CFD. Finally, these computationally optimized bioreactor-scaffold assemblies are additively manufactured and cultured with pre-osteoblast cells for 7, 20, and 24 days to achieve tissue growth in the shape of human calcaneus bones of 13 mL volume and 1 cm thickness. This framework enables an intelligent model-based design of 3D printed scaffolds and perfusion bioreactors which enhances nutrient transport for long-term volumetric tissue growth in personalized implant shapes. The novel methods described here are readily applicable for use with different cell types, biomaterials, and scaffold microstructures to research therapeutic solutions for a wide range of tissues.

Optimizing Mineralization of Bioprinted Bone Utilizing Type-2 Fuzzy Systems

Bioprinting is an emerging tissue engineering method used to generate cell-laden scaffolds with high spatial resolution. Bioprinting parameters, such as pressure, nozzle size, and speed, highly influence the quality of the bioprinted construct. Moreover, cell suspension density and other critical biological parameters directly impact the biological function. Therefore, an approximation model that can be used to find the optimal bioprinting parameter settings for bioprinted constructs is highly desirable. Here, we propose a type-2 fuzzy model to handle the uncertainty and imprecision in the approximation model. Specifically, we focus on the biological parameters, such as the culture period, that can be used to maximize the output value (mineralization volume 21.8 mm3 with the same culture period of 21 days). We have also implemented a type-1 fuzzy model and compared the results with the proposed type-2 fuzzy model using two levels of uncertainty. We hypothesize that the type-2 fuzzy model may be preferred in biological systems due to the inherent vagueness and imprecision of the input data. Our numerical results confirm this hypothesis. More specifically, the type-2 fuzzy model with a high uncertainty boundary (30%) is superior to type-1 and type-2 fuzzy systems with low uncertainty boundaries in the overall output approximation error for bone bioprinting inputs.

Achieving robustness to aleatoric uncertainty with heteroscedastic Bayesian optimisation

Bayesian optimisation is a sample-efficient search methodology that holds great promise for accelerating drug and materials discovery programs. A frequently-overlooked modelling consideration in Bayesian optimisation strategies however, is the representation of heteroscedastic aleatoric uncertainty. In many practical applications it is desirable to identify inputs with low aleatoric noise, an example of which might be a material composition which displays robust properties in response to a noisy fabrication process. In this paper, we propose a heteroscedastic Bayesian optimisation scheme capable of representing and minimising aleatoric noise across the input space. Our scheme employs a heteroscedastic Gaussian process surrogate model in conjunction with two straightforward adaptations of existing acquisition functions. First, we extend the augmented expected improvement heuristic to the heteroscedastic setting and second, we introduce the aleatoric noise-penalised expected improvement (ANPEI) heuristic. Both methodologies are capable of penalising aleatoric noise in the suggestions. In particular, the ANPEI acquisition yields improved performance relative to homoscedastic Bayesian optimisation and random sampling on toy problems as well as on two real-world scientific datasets. Code is available at: https://github.com/Ryan-Rhys/Heteroscedastic-BO

A spatiotemporal microenvironment model to improve design of a 3D bioreactor for red cell production

Cellular microenvironments provide stimuli including paracrine and autocrine growth factors and physico-chemical cues, which support efficient in vivo cell production unmatched by current in vitro biomanufacturing platforms. While three-dimensional (3D) culture systems aim to recapitulate niche architecture and function of the target tissue/organ, they are limited in accessing spatiotemporal information to evaluate and optimize in situ cell/tissue process development. Herein, a mathematical modelling framework is parameterized by single-cell phenotypic imaging and multiplexed biochemical assays to simulate the non-uniform tissue distribution of nutrients/metabolites and growth factors in cell niche environments. This model is applied to a bone marrow mimicry 3D perfusion bioreactor containing dense stromal and hematopoietic tissue with limited red blood cell (RBC) egress. The model characterized an imbalance between endogenous cytokine production and nutrient starvation within the microenvironmental niches, and recommended increased cell inoculum density and enhanced medium exchange, guiding the development of a miniaturized prototype bioreactor. The second-generation prototype improved the distribution of nutrients and growth factors and supported a 50-fold increase in RBC production efficiency. This image-informed bioprocess modelling framework leverages spatiotemporal niche information to enhance biochemical factor utilization and improve cell manufacturing in 3D systems.

Modelling for Digital Twins—Potential Role of Surrogate Models

The application of white box models in digital twins is often hindered by missing knowledge, uncertain information and computational difficulties. Our aim was to overview the difficulties and challenges regarding the modelling aspects of digital twin applications and to explore the fields where surrogate models can be utilised advantageously. In this sense, the paper discusses what types of surrogate models are suitable for different practical problems as well as introduces the appropriate techniques for building and using these models. A number of examples of digital twin applications from both continuous processes and discrete manufacturing are presented to underline the potentials of utilising surrogate models. The surrogate models and model-building methods are categorised according to the area of applications. The importance of keeping these models up to date through their whole model life cycle is also highlighted. An industrial case study is also presented to demonstrate the applicability of the concept.