Scm6A: A Fast and Low-cost Method for Quantifying m6A Modifications at the Single-cell Level.

Area of Science:

• * Molecular Biology
• * Bioinformatics
• * Computational Biology

Background:

• * N6-methyladenosine (m6A) is crucial in gene regulation but shows intercellular specificity, complicating detection.
• * Existing m6A quantification methods struggle with single-cell resolution.
• * Understanding single-cell m6A dynamics is vital for deciphering cellular heterogeneity and disease mechanisms.

Purpose of the Study:

• * To introduce Single-cell m6A Analysis (Scm6A), a novel machine learning-based method for single-cell m6A quantification.
• * To validate the accuracy and reliability of Scm6A using independent experimental data.
• * To explore the landscape and regulatory mechanisms of m6A in T cell subtypes within disease contexts.

Main Methods:

• * Developed Scm6A, a machine learning model utilizing m6A trans regulator expression and cis sequence features.
• * Applied Scm6A to single-cell RNA sequencing (scRNA-seq) data from PBMCs, lung cancer, and COVID-19 blood samples.
• * Validated Scm6A predictions by comparing with N6-methyladenosine sequencing (m6A-seq) data from magnetically sorted CD4+ and CD8+ T cells.

Main Results:

• * Scm6A demonstrated high prediction efficiency and reliability for single-cell m6A quantification.
• * Calculated m6A levels by Scm6A showed a significant positive correlation with m6A-seq results.
• * Revealed the m6A landscape and regulatory mechanisms in T cell subtypes from lung cancer and COVID-19 patients.

Conclusions:

• * Scm6A is a novel, dependable, and accurate method for single-cell m6A detection.
• * The method facilitates the study of m6A dynamics at single-cell resolution.
• * Scm6A has broad applications in m6A-related research, particularly in understanding disease states.