SEECER

SEECER corrects sequencing errors in RNA-Seq reads to improve transcriptome assembly and read alignment by modeling RNA-specific error patterns.

Key Features:

• HMM-based modeling: SEECER uses a hidden Markov Model (HMM) framework to model sequencing errors in RNA-Seq data.
• Large-scale HMM learning: It learns from hundreds of thousands of HMMs to capture sequence-specific error profiles.
• RNA-specific complexity handling: SEECER accounts for non-uniform transcript abundance, polymorphisms, and alternative splicing.
• Error identification and correction: The method identifies and corrects sequencing errors within RNA-Seq reads.
• Improved alignment: Empirical evaluations on human RNA-Seq data show improved read alignment accuracy to the genome.
• Enhanced assembly precision: SEECER enhances de novo transcriptome assembly precision compared to previous methods.
• Reference-free applicability: The approach operates in de novo transcriptome studies where no reference genome is available.
• Facilitates novel discovery: SEECER-enabled corrections have facilitated discovery of novel transcripts, including findings in Parastichopus parvimensis.

Scientific Applications:

• Read alignment improvement: Improving read alignment accuracy for human RNA-Seq datasets.
• De novo transcriptome assembly: De novo transcriptome assembly and analysis in non-model organisms such as Parastichopus parvimensis.
• Novel transcript discovery: Comparative transcriptomics and discovery of novel transcripts with subsequent experimental validation.
• Developmental transcriptomics: Analysis of transcriptomic changes across developmental stages in studied organisms.

Methodology:

SEECER employs a hidden Markov Model (HMM) framework and learns from hundreds of thousands of HMMs to identify and correct sequencing errors in RNA-Seq reads while accounting for RNA-specific error patterns.