ToxCodAn

ToxCodAn improves annotation of toxin coding sequences in transcriptome assemblies, focusing on snake venom gland transcriptomes to enable accurate identification of toxin genes.

Key Features:

• Implementation: Implemented as a Python script for processing transcriptome assemblies.
• Target data: Operates on transcriptome assemblies derived from next-generation sequencing transcriptomics, with emphasis on venom gland transcriptomes in snakes.
• Toxin coding sequence identification: Identifies coding sequences (CDS) corresponding to toxin genes within transcriptome assemblies.
• Runtime performance: Achieves run time improvements reported as more than 20 times faster than comparator annotation software.
• Coding sequence accuracy: Produces coding sequence predictions with reported accuracy over three times greater than alternatives.
• False positive reduction: Reduces incorrect toxin predictions by more than fourfold compared to other annotators.
• Validation and benchmarking: Validated using previously curated transcriptomes and compared against other annotation tools.
• Case study application: Applied in a case study on Bothrops alternatus venom gland transcriptomes.
• Adaptability: Can be extended to support toxin annotation and novel toxin detection across other venomous lineages.

Scientific Applications:

• Venom gene annotation: Generation of accurate toxin gene annotations in snake venom gland transcriptomes.
• Comparative venom evolution: Support for transcriptomics-based studies of venom evolution across venomous lineages.
• Novel toxin discovery: Identification of candidate novel toxin coding sequences across diverse species.
• Pipeline benchmarking: Use as a benchmark for comparing annotation accuracy and false positive rates among annotators.

Methodology:

Implemented as a Python script applied to transcriptome assemblies (venom gland transcriptomes) and validated by comparison against other annotators using previously curated transcriptomes and a Bothrops alternatus case study.