Fryer, T. J. | bioRxiv Science

bioRxiv2023

Machine-learning guided Venom Induced Dermonecrosis Analysis tooL: VIDAL

Snakebite envenoming is a global public health issue that causes significant morbidity and mortality, particularly in low-income regions of the world. The clinical manifestations of envenomings vary depending on the snakes venom, with paralysis, haemorrhage, and necrosis being the most common and medically relevant effects. To assess the efficacy of antivenoms against dermonecrosis, a preclinical testing approach involves in vivo mouse models that mimic local tissue effects of cytotoxic snakebites in humans. However, current methods for assessing necrosis severity are time-consuming and susceptible to human error. To address this, we present the Venom Induced Dermonecrosis Analysis tool (VIDAL), a machine-learning-guided image-based solution that can automatically identify dermonecrotic lesions in mice, adjust for lighting biases, scale the image, extract lesion area and discolouration, and calculate the severity of dermonecrosis. We also introduce a new unit, the dermonecrotic unit (DnU), to better capture the complexity of dermonecrosis severity. Our tool is comparable to the performance of state-of-the-art histopathological analysis, making it an accessible, accurate, and reproducible method for assessing dermonecrosis. Given the urgent need to address the neglected tropical disease that is snakebite, high-throughput technologies such as VIDAL are crucial in developing and validating new and existing therapeutics for this debilitating disease.

pharmacology and toxicology↗

bioRxiv2023

A Comparative Study of Protein Structure Prediction Tools for Challenging Targets: Snake Venom Toxins

Protein structure determination is a critical aspect of biological research, enabling us to understand protein function and potential applications. Recent advances in deep learning and artificial intelligence have led to the development of several protein structure prediction tools, such as AlphaFold2 and ColabFold. However, their performance has primarily been evaluated on well-characterised proteins, and comparisons using proteins with poor reference templates are lacking. In this study, we evaluated three modelling tools on their prediction of over 1000 snake venom toxin structures with no reference templates. Our findings show that AlphaFold2 (AF2) performed the best across all assessed parameters. We also observed that ColabFold (CF) only scored slightly worse than AF2, while being computationally less intensive. All tools struggled with regions of intrinsic disorder, such as loops and propeptide regions, and performed well in predicting the structure of functional domains. Overall, our study highlights the importance of exercising caution when working with proteins that have poor reference templates, are large, and contain flexible regions. Nonetheless, leveraging computational structure prediction tools can provide valuable insights into the modelling of protein interactions with different targets and reveal potential binding sites, active sites, and conformational changes, as well as into the design of potential molecular binders for reagent, diagnostic, or therapeutic purposes. StatementRecent advances in machine learning have led to the development of new protein structure prediction tools. However, these tools have mainly been tested on well-known proteins and their performance on proteins without known templates is unclear. This study evaluated the performance of three tools on over 1000 snake venom toxins. We found that while caution is required when studying poorly characterised proteins, these tools offer valuable opportunities to understand protein function and applications.

bioinformatics↗