Abstract

Typically, most omics analysis (proteomic and transcriptomic) of snakes are focused on the dominant enzymatic proteins used for evolutionary analysis or those engaged in envenoming symptoms. This study presents a comprehensive multi-assembler transcriptomic analysis focused on the non-dominant and enzymatic or non-enzymatic putative proteins of the venom glands of three medically significant Colombian snake species. Together, these results highlight how continued improvements in modern omics workflows, coupled with extensive manual curation, enable more complete putative protein variants discovery when multiple assemblers are integrated. Here, we reconstructed the toxinomes of the viperids Bothrops asper and Crotalus durissus cumanensis, and the elapid Micrurus mipartitus, by comparing four assemblers (Trinity, SPAdes, SOAPdenovo-Trans k = 31 and k = 97) and integrating them into a non-redundant meta-assembly. Protein-candidate alignments were extensively inspected, and validation of conserved domains and functional motifs are discussed. The curated toxinomes revealed substantial diversity across major and accessory families, and assembler choice strongly affected transcript variant recovery. Together, these results provide a more comprehensive view of venom-gland transcriptome analysis and diversity, expanding the set of candidate venom components for future functional and proteomic validation, with potential implications for venom composition studies and antivenom development.