msup6 supA

msup6 supA predicts m6A methylation status and the effects of trait-associated SNPs in maize RNA sequences using a weakly supervised learning approach trained on low-resolution epitranscriptome data such as MeRIP-seq.

Key Features:

• Weakly Supervised Learning Model: A weakly supervised model trained on low-resolution epitranscriptome datasets (e.g., MeRIP-seq) predicts m6A methylation status across maize RNA fragments and genomic regions.
• SNP Trait Association: Identification and prediction of single nucleotide polymorphisms (SNPs) that can add or remove m6A modifications, linking SNPs to potential epitranscriptomic regulatory mechanisms and trait associations.
• Experimentally Validated m6A Regions: Includes a consolidated dataset of 58,838 experimentally validated m6A-containing regions in maize used for training and testing.
• Predicted Trait-Associated SNPs: Contains 2,578 predicted trait-associated SNPs reported to affect m6A modifications and their predicted impacts on methylation status.
• Model-Based Site Prediction: Computational prediction of putative m6A sites from maize sequences and assessment of how specific SNPs alter predicted m6A status.

Scientific Applications:

• m6A mapping in maize: Predicts nucleotide-resolution m6A methylation patterns from low-resolution MeRIP-seq-derived training data to inform epitranscriptome studies in maize.
• SNP impact analysis: Identifies SNPs that potentially create or abolish m6A sites to study genetic regulation of RNA methylation and associated phenotypic traits.
• Dataset provision for modeling: Provides a curated set of 58,838 validated m6A regions and associated training/testing data for benchmarking and developing m6A prediction algorithms.
• Mechanistic inference: Enables investigation of how epitranscriptomic variation mediated by SNPs may contribute to regulatory mechanisms in maize.

Methodology:

Integration of weakly supervised learning techniques with low-resolution epitranscriptome sequencing data (e.g., MeRIP-seq) to predict m6A methylation patterns and infer SNP effects on m6A status.