Publications

Biofilms are critical for understanding environmental processes, developing biotechnology applications, and progressing in medical treatments of various infections. Nowadays, a key limiting factor for biofilm analysis is the difficulty in obtaining large datasets with fully annotated images. This study introduces a versatile approach for creating synthetic datasets of annotated biofilm images with employing deep generative modeling techniques, including VAEs, GANs, diffusion models, and CycleGAN. Synthetic datasets can significantly improve the training of computer vision models for automated biofilm analysis, as demonstrated with the application of Mask R-CNN detection model. The approach represents a key advance in the field of biofilm research, offering a scalable solution for generating high-quality training data and working with different strains of microorganisms at different stages of formation. Terabyte-scale datasets can be easily generated on personal computers. A web application is provided for the on-demand generation of biofilm images.

Adapting biological systems for nanoparticle synthesis opens an orthogonal Green direction in nanoscience by reducing the reliance on harsh chemicals and energy-intensive procedures. This study addresses the challenge of efficient catalyst preparation for organic synthesis, focusing on the rapid formation of palladium (Pd) nanoparticles using bacterial cells as a renewable and eco-friendly support. The preparation of catalytically active nanoparticles on the bacterium Paracoccus yeei represents a more suitable approach to increase the reaction efficiency due to its resistance to metal salts. We introduce an efficient method that significantly reduces the preparation time of Pd nanoparticles on Paracoccus yeei VKM B-3302 bacteria to only 7 min, greatly accelerating the process compared with traditional methods. Our findings reveal the major role of live bacterial cells in the formation and stabilization of Pd nanoparticles, which exhibit high catalytic activity in the Mizoroki–Heck reaction. This method not only ensures high yields of the desired product but also offers a greener and more sustainable alternative to conventional catalytic processes. The rapid preparation and high efficiency of this biohybrid catalyst opens new perspectives for the application of biosupported nanoparticles in organic synthesis and a transformative sustainable pathway for chemical production processes.

Palladium catalysts form a cornerstone of modern chemistry with upmost scientific and industrial impact. Bulk palladium metal itself is chemically inert, and a sequence of chemical transformations has to be utilized to convert the metal into Pd pre-catalyst covered by ligands. However, the "cocktail" of catalysis concept discovered recently has shown that Pd systems can efficiently operate in catalysis without the necessity of a complicated and expensive pre-installed ligand environment. Here, we point out on a green and sustainable process for Pd active species generation without the need of waste-abundant pre-catalyst-related chemistry. In this work, an electric current was used to generate an active Pd catalyst from a bulk metal in an ionic liquid medium for the efficient cross-coupling of aryl iodides/bromides and boronic acids. Synthetically important Suzuki cross-coupling was utilized as a representative test reaction to confirm the idea. It should be emphasized that electric current is used only at the Pd dissolution stage. Afterwards, the electrodes are removed from the reaction mixture and a standard reaction procedure can be followed. The reported catalyst preparation process via electrochemical dissolution is potentially compatible with a number of already existing catalytic methods.