Exact Bayesian inference for the detection of graft-mobile transcripts from sequencing data

Area of Science:

• Plant molecular biology
• Genetics
• Bioinformatics

Background:

• Long-distance transport of messenger RNAs (mRNAs) is crucial for plant development.
• RNA sequencing (RNA-Seq) on grafted plants is a common method to study mobile mRNAs.
• Accurate assignment of sequenced mRNAs to their genetic origin, often via single-nucleotide polymorphisms (SNPs), is challenging due to sequencing errors.

Purpose of the Study:

• To develop a robust computational framework for identifying graft-mobile transcripts.
• To accurately distinguish true single-nucleotide polymorphisms (SNPs) from RNA sequencing errors.
• To provide guidelines for experimental design in graft-mobile mRNA detection.

Main Methods:

• Analytical computation of Bayes factors using RNA-Seq data across all SNPs within an mRNA.
• Performance evaluation through simulations to assess the accuracy of the proposed framework.
• Comparison with existing detection methods, considering variability in read depth, error rates, and multiple SNPs per transcript.

Main Results:

• Bayes factors accurately identify graft-mobile transcripts.
• The proposed framework effectively distinguishes SNPs from sequencing errors.
• Ignoring variability in read depth, error rates, and multiple SNPs can lead to misclassification.

Conclusions:

• The Bayes factor approach offers a reliable method for detecting graft-mobile mRNAs.
• Accurate identification requires considering multiple factors beyond single SNPs.
• The study provides essential criteria for successful experimental design in this field.