Nanotechnology based on nanoscale materials is rapidly being used in clinical settings, particularly as a new approach for infectious illnesses. Recently, many physical/chemical approaches utilized to produce nanoparticles are expensive and highly unsafe to biological species and ecosystems. This study demonstrated an environmentally friendly mode of producing nanoparticles (NPs) where Fusarium oxysporum has been employed for generation of silver nanoparticles (AgNPs), which were further tested for their antimicrobial potentials against a variety of pathogenic microorganisms. The characterization of NPs was completed by UV–Vis spectroscopy, DLS and TEM, where it has been found that the NPs were mostly globular, with the size range of 50 to 100 nm. The myco-synthesized AgNPs showed prominent antibacterial potency observed as zone of inhibition of 2.6 mm, 1.8 mm, 1.5 mm, and 1.8 mm against Vibrio cholerae, Streptococcus pneumoniae, Klebsiella pneumoniae and Bacillus anthracis, respectively, at 100 µM. Similarly, at 200 µM for A. alternata, A. flavus and Trichoderma have shown zone of inhibition as 2.6 mm, 2.4 mm, and 2.1 mm, respectively. Moreover, SEM analysis of A. alternata confirmed the hyphal damage where the layers of membranes were torn off, and further EDX data analysis showed the presence of silver NPs, which might be responsible for hyphal damage. The potency of NPs may be related with the capping of fungal proteins that are produced extracellularly. Thus, these AgNPs may be used against pathogenic microbes and play a beneficial role against multi-drug resistance.

Tuberculosis (TB) is one of the leading causes of death throughout the world. Traditional medicine poses a serious challenge to drug resistance against Mycobacterium tuberculosis. Hence, it is an urgent need to develop, novel the anti-mycobacterial compounds. In this study, a potent anti-mycobacterial compound producing Staphylococcus hominis IS2 was isolated from the custard apple exhibiting novel anti-mycobacterial activity than other isolates. The bacterial strain was cultured and anti-mycobacterial compound was purified by the combination of ammonium sulphate precipitation and DEAE-cellulose column chromatography. The purified protein corresponding to 10 kDa presented anti-mycobacterial activity. It was further characterized and the difference in activity towards temperature and pH was observed (p < 0.01). The purified protein was analyzed using FT-IR and MALDI-TOF MS/MS analysis. Enzymatic treatment revealed the absence of carbohydrate moieties in the purified protein. The purified anti-mycobacterial protein presented anti-cancer activity against cervical cancer cell lines (HeLa) and was dose-dependent (p < 0.01). Haemolytic assay and Zebra fish embryo studies revealed that the isolated protein is safe at lower doses.

Machine learning has become a central framework for analyzing high-dimensional biological data generated by genomics, transcriptomics, proteomics, single-cell sequencing, and multi-omics technologies. However, predictive modeling is frequently limited by severe class imbalance, where biologically important populations including rare disease subtypes, uncommon molecular states, and low-frequency cell populations are underrepresented. Synthetic oversampling addresses this challenge by increasing minority-class representation, with the Synthetic Minority Oversampling Technique (SMOTE) being the most widely used approach. However, conventional SMOTE assumes that geometric proximity reflects biological similarity, an assumption often violated in high-dimensional, nonlinear, and heterogeneous biological data. This review examines clustering-guided synthetic generation methods that integrate unsupervised learning before oversampling to preserve underlying biological structure. We evaluate partition-based, density-aware, fuzzy, cluster-filtering, and representation-based approaches, emphasizing their mathematical foundations, computational assumptions, and applications in molecular classification, rare population analysis, single-cell modeling, and multi-omics integration. Key challenges including cluster instability, preservation of rare biological populations, synthetic data validation, and distinguishing statistical rarity from biological significance are discussed. Finally, we outline emerging directions involving biological foundation models, graph-based learning, multi-omics representation spaces, and biologically constrained generative frameworks. These advances represent a shift from conventional data balancing toward structure-aware synthetic modeling that preserves meaningful biological organization while improving predictive performance.