Researchers at the International Institute of Molecular and Cell Biology in Warsaw have developed a powerful new computational tool called Ninetails to analyze mRNA poly(A) tails with unprecedented accuracy. This innovation, based on neural networks, helps scientists detect non-adenosine modifications in mRNA tails—an area previously difficult to study using existing methods.

Poly(A) tails are crucial for mRNA stability and translation, impacting both natural and synthetic mRNAs, including those used in vaccines. While direct RNA sequencing allows for single-molecule measurements of poly(A) tail length, it has lacked the capability to identify mixed compositions of nucleotides within the tails. Ninetails fills this gap by pinpointing and quantifying these modifications across different biological contexts, such as cell types, species, and even mRNA therapeutics.

A rationale for the applied signal transformation strategy. Signals that mirror each other (e.g. G and C/U) produce identical GASFs but different GADFs. 

Using Ninetails, the research team discovered that composite poly(A) tails are more common than previously thought. They found that cytoplasmic TENT5 polymerases and mitochondrially encoded mRNAs are especially rich in these modifications. Importantly, for mRNA therapeutics, they showed that poly(A) tail composition changes over time inside cells and is influenced by how synthetic mRNA is manufactured—insights that could help optimize future mRNA vaccine production.

This study highlights how advanced computational tools like Ninetails are driving a deeper understanding of RNA biology and improving biopharmaceutical development.

Availability – The source code of the Ninetails software used for data analysis is publicly available: deposited in Github [https://github.com/LRB-IIMCB/ninetails]