In silico prediction of effector proteins in oil palm pest, Metisa plana

Abstract

Metisa plana is one of the main oil palm leaf defoliators in Malaysia. Infestation of M. plana at the oil palm plantation caused severe yield loss in the past. Many control methods have been introduced to combat the infestation, but the problem persists till today. More information regarding the bagworm is needed to understand the pest-host interaction so that an effective control method can be properly designed. This study was carried out to predict the repertoire of effector proteins secreted by M. plana that could potentially suppress the host defence mechanisms. In silico screening of the secretory proteins from predicted genes was done by using several tools namely SignalP, TMHMM, TargetP and WOLFPSORT. Then, the acquired dataset was subjected to functional annotation through a similarity search against UniProt-Trembl database, gene ontology (GO) using BLAST2GO, InterPro protein domain classification, cluster of orthologous groups (COG), and KEGG pathway prediction. To determine the expression of the predicted genes, TPM value was calculated using the available transcriptomic dataset of M. plana. We identified a total of 948 secretory proteins from the initial screening of the 28,738 proteins predicted from BRAKER2 pipeline. Several proteins were classified as specific protein classes associated with effector functions in insect pests, namely detoxifying enzymes, lipases, peroxidases, Ca²⁺ binding proteins, and chemosensory and odorant binding proteins. This predicted effector data can be used as a reference for future works in validating the protein’s interaction and mechanisms with the oil palm defence responses.