Combining Group-Contribution Concept and Graph Neural Networks Toward Interpretable Molecular Property Models.

Quantitative structure-property relationships (QSPRs) are important tools to facilitate and accelerate the discovery of compounds with desired properties. While many QSPRs have been developed, they are associated with various shortcomings such as a lack of generalizability and modest accuracy. Albeit various machine-learning and deep-learning techniques have been integrated into such models, another shortcoming has emerged in the form of a lack of transparency and interpretability of such models. In this work, two interpretable graph neural network (GNN) models (attentive group-contribution (AGC) and group-contribution-based graph attention (GroupGAT)) are developed by integrating fundamentals using the concept of group contributions (GC). The interpretability consists of highlighting the substructure with the highest attention weights in the latent representation of the molecules using the attention mechanism. The proposed models showcased better performance compared to classical group-contribution models, as well as against various other GNN models describing the aqueous solubility, melting point, and enthalpies of formation, combustion, and fusion of organic compounds. The insights provided are consistent with insights obtained from the semiempirical GC models confirming that the proposed framework allows highlighting the important substructures of the molecules for a specific property.

Variables and Mechanisms Affecting Electro-Membrane Extraction of Bio-Succinic Acid from Fermentation Broth.

The production of succinic acid from fermentation is a promising approach for obtaining building-block chemicals from renewable sources. However, the limited bio-succinic yield from fermentation and the complexity of purification has been making the bio-succinic acid production not competitive with petroleum-based succinic acid. Membrane electrolysis has been identified to be a promising technology in both production and separation stages of fermentation processes. This work focuses on identifying the key operational parameters affecting the performance of the electrolytic cell for separating succinic acid from fermentation broth through an anionic exchange membrane. Indeed, while efforts are mainly focused on studying the performance of an integrated fermenter-electrolytic cell system, a lack of understanding remains in how to tune the electrolytic cell and which main parameters are involved. The results show that a single electrolytic cell of operating volume 250 mL was able to extract up to 3 g L-1 h-1 of succinic acid. The production of OH- ions by water electrolysis can act as a buffer for the fermenter and it could be tuned as a function of the extraction rate. Furthermore, as the complexity of the solution in terms of the quantity and composition of the ions increased, the energy required for the separation process decreased.

Towards the Development of Digital Twins for the Bio-manufacturing Industry.

The bio-manufacturing industry, along with other process industries, now has the opportunity to be engaged in the latest industrial revolution, also known as Industry 4.0. To successfully accomplish this, a physical-to-digital-to-physical information loop should be carefully developed. One way to achieve this is, for example, through the implementation of digital twins (DTs), which are virtual copies of the processes. Therefore, in this paper, the focus is on understanding the needs and challenges faced by the bio-manufacturing industry when dealing with this digitalized paradigm. To do so, two major building blocks of a DT, data and models, are highlighted and discussed. Hence, firstly, data and their characteristics and collection strategies are examined as well as new methods and tools for data processing. Secondly, modelling approaches and their potential of being used in DTs are reviewed. Finally, we share our vision with regard to the use of DTs in the bio-manufacturing industry aiming at bringing the DT a step closer to its full potential and realization.

Towards smart biomanufacturing: a perspective on recent developments in industrial measurement and monitoring technologies for bio-based production processes.

The biomanufacturing industry has now the opportunity to upgrade its production processes to be in harmony with the latest industrial revolution. Technology creates capabilities that enable smart manufacturing while still complying with unfolding regulations. However, many biomanufacturing companies, especially in the biopharma sector, still have a long way to go to fully benefit from smart manufacturing as they first need to transition their current operations to an information-driven future. One of the most significant obstacles towards the implementation of smart biomanufacturing is the collection of large sets of relevant data. Therefore, in this work, we both summarize the advances that have been made to date with regards to the monitoring and control of bioprocesses, and highlight some of the key technologies that have the potential to contribute to gathering big data. Empowering the current biomanufacturing industry to transition to Industry 4.0 operations allows for improved productivity through information-driven automation, not only by developing infrastructure, but also by introducing more advanced monitoring and control strategies.