Dr. Margaux Aubel, Okinawa Institute of Science and Technology

Detecting Novel Programmed Ribosomal Frameshift Sites using Fourier Transformed Rates of Evolution

Programmed ribosomal frameshifting (PRF) occurs when ribosomes switch the reading frame while translating mRNA at a specific position and at a defined rate. While PRF is a well-characterized strategy in viruses to regulate protein ratios, evidence for PRF in cellular organisms remains sparse. The best-studied example is the antizyme gene in yeast and mammals, where a conserved +1 frameshift is essential to translate the complete functional protein. Due to the limited number of known PRF genes, especially in non-viral sequences, no general predictor of PRF currently exists. Here, we present an evolutionary sequence-based framework to predict PRF using codon-wise periodic nucleotide conservation. Coding sequences exhibit a characteristic periodicity in nucleotide conservation, reflecting the different selective constraints across codon positions. We quantify this periodic signal using Fourier transformation and infer the most likely reading frame along a sequence to identify and predict candidate PRF events. We subsequently evaluate performance on well-characterized coding sequences with experimentally verified reading frames, demonstrating the predictor's ability to recover known PRF genes. Finally, we apply the predictor on yeast coding sequences to identify new PRF candidates that we will verify experimentally.

Weitere Infos zur Veranstaltung