Biology subjects

KAJAVA, A. V.

Publications and source records attributed to KAJAVA, A. V..

5 recordsLinked to original sources

bioRxiv2026

LLPSight: enhancing prediction of LLPS-driving proteins using machine learning and protein Language Models

In eukaryotic cells, essential functions are often confined within organelles enclosed by lipid membranes. Increasing evidence, however, highlights the role of membrane-less organelles (MLOs), formed through liquid-liquid phase separation (LLPS). MLO assemblies are typically initiated by "driver" proteins, which form a scaffold to recruit additional "client" molecules. By leveraging expanding MLO datasets and modern machine learning approaches, we developed LLPSight, an ML-based predictor of LLPS-driving proteins. The model was trained using rigorously curated datasets: a positive set of proteins experimentally confirmed to drive LLPS in vivo and a negative set of soluble, unstructured proteins not associated with LLPS. For the features, we employed a cutting-edge approach using embeddings from protein Language Models. LLPSight achieves the highest F1 score (0.885) among existing tools, enabling more efficient discovery of new LLPS drivers eagerly awaited by researchers for experimental validation. An additional key feature of LLPSight is its ability to perform proteome-wide analyses; application to the human proteome yielded promising targets. LLPSight can be obtained from authors upon request.

bioinformatics↗

bioRxiv2024

Frameshifts in Tandem Repeats: Consequences on Protein Physicochemical Properties and Function

The genetic code uses three-nucleotide units to encode each amino acid in proteins. Insertions or deletions of nucleotides not divisible by three shift the reading frames, resulting in significantly different protein sequences. These events are disruptive but can also create variability important for evolution. Previous studies suggest that genetic code and gene sequences evolve to minimize frameshift effects, maintaining similar physicochemical properties to their reference proteins. Here, we focused on tandem repeat sequences, known as frameshift hotspots. Using cutting-edge bioinformatics tools, we compared reference and frameshifted protein sequences within tandem repeats across 50 prokaryotic and eukaryotic proteomes. Our analysis revealed several intriguing sequence-structure-function correlations. We showed that in contrast to the general tendency, frameshifts within these regions, especially with short repeats, lead to significant changes: increased hydrophobicity and arginine content, new aggregation-prone and transmembrane regions. Overall, frameshifts have stronger effects on tandem repeat regions compared to non-repetitive sequences, and therefore can be a primary cause of altered functions, cellular localization, and the development of various pathologies.

bioinformatics↗

bioRxiv2024

Quantitative model and physical mechanisms of iRBC membrane curling during egress of malaria parasites

Egress of malaria merozoites from infected red blood cells (iRBC) is a critical step in the parasites life cycle. The egress is accompanied by the formation of a pore in the erythrocyte membrane, followed by an outward curling of the membrane around the pore, resulting in a complete eversion of the erythrocyte membrane, pushing the parasites away. Despite the well-documented steps of the egress, the detailed mechanism and source of energy for such a spectacular eversion of iRBC remains largely unknown. In this paper, we consider a biophysical model based on the energetics of the egress process that includes both viscous dissipation and energy consumption for the formation of the rim around the pore in iRBC. We show that viscosity does not play a significant role in iRBC eversion and we hypothesize that this process is controlled by lateral lipid diffusion. The model is supported by quantitative estimates and is in good agreement with known experimental data.

biophysics↗

bioRxiv2022

Census of exposed aggregation-prone regions in proteomes

Loss of solubility usually leads to the detrimental elimination of protein function. In some cases, the protein aggregation is also required for beneficial functions. Given the duality of this phenomenon, it remains a fundamental question how natural selection controls the aggregation. The exponential growth of genomic sequence data and recent progress with in silico predictors of the aggregation allows approaching this problem by a large-scale bioinformatics analysis. Most of the aggregation-prone regions are hidden within the 3D structures and, therefore, they cannot realize their potential to aggregate. Thus, the most realistic census of the aggregation prone regions requires crossing aggregation prediction with information about the location of the natively unfolded regions. This allows us to detect so-called "Exposed Aggregation-prone Regions" (EARs). Here, we analyzed the occurrence and distribution of the EARs in 76 full reference proteomes from the three kingdoms of life. For this purpose, we used a bioinformatics pipeline, which provides a consensual result based on several predictors of aggregation. Our analysis revealed a number of new statistically significant correlations about the presence of EARs in different organisms, their dependence on protein length, cellular localizations, co-occurrence with short linear motifs, and the level of protein expression. We also obtained a list of proteins with the conserved aggregation-prone sequences for further experimental tests. Insights gained from this work led to a deeper understanding of the functional and evolutionary relations of the protein aggregation.

molecular biology↗

bioRxiv2020

Trimeric SARS-CoV-2 Spike proteins produced from CHO-cells in bioreactors are high-quality antigens

The Spike protein of SARS-CoV-2 is essential for virus entry into human cells. In fact, most neutralizing antibodies against SARS-CoV-2 are directed against the Spike, making it the antigen of choice for use in vaccines and diagnostic tests. In the current pandemic context, global demand for Spike proteins has rapidly increased and could exceed hundreds of grams to kilograms annually. Coronavirus Spikes are large, heavily glycosylated, homotrimeric complexes, with inherent instability. Their poor manufacturability now threatens availability of these proteins for vaccines and diagnostic tests. Here, we outline a scalable, GMP-compliant, chemically defined process for production of a cell secreted, stabilized form of the trimeric Spike protein. The process is chemically defined and based on clonal, suspension-CHO cell populations and on protein purification via a two-step, scalable downstream process. The trimeric conformation was confirmed using electron microscopy and HPLC analysis. Binding to susceptible cells was shown using a virus-inhibition assay. The diagnostic sensitivity and specificity for detection of serum SARS-CoV-2 specific IgG1 was investigated and found to exceed that of Spike fragments (S1 and RBD). The process described here will enable production of sufficient high-quality trimeric Spike protein to meet the global demand for SARS-CoV-2 vaccines and diagnostic tests.

cell biology↗