ClusTrast: a short read de novo transcript isoform assembler guided by clustered contigs

Area of Science:

• Bioinformatics
• Genomics
• Computational Biology

Background:

• De novo transcriptome assembly is crucial for species without reference genomes.
• Existing de novo methods struggle with accurate reconstruction of transcript isoforms.
• Comprehensive isoform reconstruction is essential for understanding gene expression.

Purpose of the Study:

• To develop an improved de novo transcriptome assembler for comprehensive isoform reconstruction.
• To address the limitations of current assemblers in capturing transcript diversity.
• To facilitate the study of transcript isoforms in non-model organisms.

Main Methods:

• Developed ClusTrast, a de novo transcript isoform assembler pipeline.
• Input: short-read RNA-seq data.
• Pipeline: primary assembly, guiding contig clustering, read alignment, cluster-wise assembly, and final assembly merging.

Main Results:

• ClusTrast identified more known isoforms than other de novo assemblers across six eukaryotic species.
• Achieved superior recall for low-expressed isoforms (<15% percentile).
• Reconstructed a high percentage of reference transcript length (35-69%) and captured polymorphic variants (58-81%).

Conclusions:

• ClusTrast is a valuable tool for comprehensive transcriptome assembly in species without reference genomes.
• Particularly useful for identifying polymorphic variants and diverse transcript isoforms.
• Enhances transcriptomic studies in non-model organisms.