In a paper published in iFungi, an international team of scientists present evidence of alternating echinocandin therapy with liposomal amphotericin B effectively clears Nakaseomyces glabratus infection and reduces the emergence of echinocandin resistance. The study reveals that targeting echinocandin tolerant fungal subpopulations before they evolve into resistant populations may provide a new strategy for preserving the Clinical application of existing antifungal echinocandin drugs.

Guided by the classic notion that a protein's amino acid sequence determines its three-dimensional structure, protein structure prediction has long been a central problem in structural biology. Over the past decade, deep learning methods, exemplified by AlphaFold2 (AF2), have revolutionized the field, an achievement recognized by the 2024 Nobel Prize in Chemistry. Yet this success does not mean the protein folding problem is fully solved. Key challenges, each an opportunity for future breakthroughs, include single-sequence prediction, modeling protein dynamics, predicting multimeric complexes, and incorporating experimental constraints.

Researchers at Fudan University, in collaboration with Zhejiang Cancer Hospital (Hangzhou Institute of Medicine, Chinese Academy of Sciences) and Washington University in St. Louis, have released bacNeo, the first comprehensive computational toolkit dedicated to identifying bacteria-derived neoantigens in human cancers. The framework combines three modules, BACC for bacterial component classification, BACH for HLA allele typing, and BACP for bacterial peptide prioritization, into a single one-stop pipeline that accommodates metagenomic, metatranscriptomic, bulk and single-cell sequencing, and mass spectrometry data. A novel scoring system, SPARK (Score for Peptide Antigenicity Recognition and Kinetics), weights TAP transport efficiency against HLA-peptide binding affinity and reaches an area under the ROC curve of 0.81 on experimentally validated peptide-HLA datasets. The work, published in iMetaMed, provides a reproducible and modular foundation for systematic bacterial neoantigen discovery and supports the design of therapeutic vaccines, chimeric antigen receptor (CAR) T-cell therapies, and immune checkpoint blockade strategies across diverse cancer types.

With rapid advances in synthetic biology and genetic engineering, genetically engineered bacteria (GEB) have emerged as a promising platform for biological therapy, addressing key limitations of conventional drug delivery systems and demonstrating significant clinical potential. By leveraging a modular design framework, GEB exhibits high biocompatibility and programmability, enabling precise genetic modifications for targeted delivery, dynamic responsiveness, and sustained therapeutic effects. This review systematically examines the therapeutic development of GEB, focusing on its pivotal roles in tumor therapy, inflammatory bowel disease (IBD), metabolic disorders, and the treatment of complex infections. It outlines key design strategies—including sensing, payload release, and biocontainment modules—as well as evaluation frameworks and mechanisms of action. We also discuss major challenges to clinical translation, such as the limited availability of clinically validated chassis strains, the balance between bacterial fitness and therapeutic load, and persistent biosafety concerns. Future progress will depend on advancements in high-precision genome editing, intelligent responsive systems, and interdisciplinary collaboration to bridge the gap between preclinical promise and clinical application. This review aims to provide an up-to-date reference to facilitate the translation of GEB from laboratory discoveries to viable clinical applications in precision medicine.

Mycobacterial kinase PknG directly phosphorylates host lysine-tRNA ligase (KARS) at T592 and T595, modulating the MAPK/NF‑κB signaling axis and regulating host immunity. Targeting these phosphorylation sites offers a new host-directed therapy against drug-resistant tuberculosis.