Unveiling hidden toxin diversity: Discovery of novel venom components through manual curation of highly expressed sequences annotated as “no hits” in Phoneutria nigriventer spider venom gland transcriptome

Abstract

Spider venoms have evolved over thousands of years, optimizing feeding and defense mechanisms. Venom components show pharmacological and biotechnological potential, rising interest in their study. However, the isolation of spider toxins for experimental evaluation poses significant challenges. To address this, transcriptomic analysis combined with computational tools has emerged as an appealing approach to characterizing spider venoms. However, many sequences remain unidentified after automatic annotation. In this study, we manually curated a subset of previously unannotated sequences from the Phoneutria nigriventer transcriptome and identified new putative venom components. Our manual analysis revealed 29 % of the analyzed sequences were potential venom components, 29 % hypothetical/uncharacterized proteins, and 17 % cellular function proteins. Only 25 % of the originally unannotated dataset remained without any identification. Most reclassified components were cysteine-rich peptides, including 23 novel putative toxins. We also found glycine-rich peptides (GRP), corroborating the previous description of GRPs in Phoneutria pertyi venom glands. Furthermore, to emphasize the recurrence of the lack of annotation in spider venom glands transcripts, we provide a survey of the percentage of unidentified sequences in several published spider venom transcriptomics studies. In conclusion, our study highlights the importance of manual curation in uncovering novel venom components and underscores the need for improved annotation strategies to fully exploit the medical and biotechnological potential of spider venoms.