Machine-learning-guided proteomic profiling reveals chitosan-modulated parasitism in the Pochonia chlamydosporia–root-knot nematode interaction

The nematophagous fungus Pochonia chlamydosporia (Pc) parasitizes Meloidogyne javanica (MELJA) eggs, and chitosan (T8) modulates this interaction. However, the molecular mechanisms underlying this tripartite system remain poorly understood. We performed an integrative proteomic analysis of fungal and nematode responses under four conditions (Pc, PcT8, MELJA_Pc, MELJA_Pc_T8) at 24 and 96 h post-inoculation (hpi), combining differential abundance, sequence–feature clustering, phylogenetic comparisons, and machine-learning approaches. Proteome-wide analyses indicated a stimulus-dependent restructuring of Pc, with multivariate models identifying early metabolic activators and late cell wall–associated enzymes as key drivers of its temporal response. The fungal secretome was structured by functional convergence rather than evolutionary ancestry, with extracellular hydrolases and CFEM proteins dominating Cellular Component predictions. Chitosan was associated with a strong divergence in the nematode proteome, suppressing vitellogenins, ZP-domain proteins, and Piwi regulators while enhancing ATP synthase and EF1-α isoforms. Functional enrichment suggested that chitosan coincided with shifts in nematode pathways related to cytoskeletal reinforcement and chromatin activity, while fungal parasitism alone induced gene-silencing and metabolic-stress signatures. Thirteen consensus biomarkers across machine-learning models captured these shifts. Chitosan is associated with enhanced fungal parasitism, coinciding with the activation of conserved secretory modules in Pc and proteomic shifts in MELJA that involve resilience-related pathways. These findings provide a machine-learning-guided framework for identifying molecular biomarkers of nematode susceptibility and fungal parasitism under controlled in vitro conditions .