Enhanced compound-protein binding affinity prediction by representing protein multimodal information via a coevolutionary strategy.

Due to the lack of a method to efficiently represent the multimodal information of a protein, including its structure and sequence information, predicting compound-protein binding affinity (CPA) still suffers from low accuracy when applying machine-learning methods. To overcome this limitation, in a novel end-to-end architecture (named FeatNN), we develop a coevolutionary strategy to jointly represent the structure and sequence features of proteins and ultimately optimize the mathematical models for predicting CPA. Furthermore, from the perspective of data-driven approach, we proposed a rational method that can utilize both high- and low-quality databases to optimize the accuracy and generalization ability of FeatNN in CPA prediction tasks. Notably, we visually interpret the feature interaction process between sequence and structure in the rationally designed architecture. As a result, FeatNN considerably outperforms the state-of-the-art (SOTA) baseline in virtual drug evaluation tasks, indicating the feasibility of this approach for practical use. FeatNN provides an outstanding method for higher CPA prediction accuracy and better generalization ability by efficiently representing multimodal information of proteins via a coevolutionary strategy.

Develop Targeted Protein Drug Carriers through a High-Throughput Screening Platform and Rational Design.

Protein-based drugs offer advantages, such as high specificity, low toxicity, and minimal side effects compared to small molecule drugs. However, delivery of proteins to target tissues or cells remains challenging due to the instability, diverse structures, charges, and molecular weights of proteins. Polymers have emerged as a leading choice for designing effective protein delivery systems, but identifying a suitable polymer for a given protein is complicated by the complexity of both proteins and polymers. To address this challenge, a fluorescence-based high-throughput screening platform called ProMatch to efficiently collect data on protein-polymer interactions, followed by in vivo and in vitro experiments with rational design is developed. Using this approach to streamline polymer selection for targeted protein delivery, candidate polymers from commercially available options are identified and a polyhexamethylene biguanide (PHMB)-based system for delivering proteins to white adipose tissue as a treatment for obesity is developed. A branched polyethylenimine (bPEI)-based system for neuron-specific protein delivery to stimulate optic nerve regeneration is also developed. The high-throughput screening methodology expedites identification of promising polymer candidates for tissue-specific protein delivery systems, thereby providing a platform to develop innovative protein-based therapeutics.

Release and removal of algal organic matter during prechlorination and coagulation treatment of cyanobacteria-laden water: Are we on track?

A better understanding of the physicochemical properties and fate of algae-derived organic matter (AOM) in water treatments significantly benefits the control of algae-derived disinfection byprodcuts and process parameter optimization. In this study, we conducted a comprehensive investigation of the release and treatability of dissolved organic matter during prechlorination and postcoagulation treatments of cyanobacteria-laden source water via size-exclusion chromatography-tandem diode array detector, fluorescence detector and organic carbon detector. The results revealed that the allochthonous humic substances could protect algal cell membrane from damage during prechlorination at a low level of chlorine dose. Due to the release and oxidation of biopterins during prechlorination of M. aeruginosa cells, the variation of the humic-like fluorescence can be used to indicate the chlorine dose for a sufficient membrane damage of algae cells. The prechlorination of M. aeruginosa cells induced minimal release of large MW biopolymer fractions but much more release of low MW fractions E1 and E2 (i.e., unknown carbonaceous substances and fluorescent nitrogenous biopterins). The physically extracted AOM contained a large proportion of biopolymers and could not well represent those released during prechlorination treatment. During coagulation, the negative effect of humic substances on the coagulant demand to achieve algae removal was more remarkable than AOM released by prechlorination. The high-MW biopolymers and humic substances can be removed over 50% by coagulation. Among the low-MW carbonaceous fractions, E1 released by prechlorination can also be effectively removed via coagulation while fractions C, D (possibly oligopeptides or secondary aromatic metabolites & low MW acids) and nitrogenous biopterins were recalcitrant to coagulation. This study highlights the differences of AOM properties between physical extraction and prechlorination and provides a basis for drinking water treatment plants to give more attention to the recalcitrant low MW fractions in coagulation when treating algae-laden source water.