Project Machine learning and nanopore sequencing to map hidden microproteins RNA isoforms; Uwe Ohler

Standard reference catalogs describe only a fraction of what the genome actually translates. Many small open reading frames (smORFs), short stretches of RNA that can be translated into microproteins, sit outside those catalogs, and some may be produced together from a single polycistronic transcript, an RNA that carries more than one protein-coding region. Which ORF a ribosome translates can also depend on the exact RNA isoform, the particular spliced version of a transcript, and conventional methods struggle to link a given translation event to the specific isoform it came from.

This project maps translation at single-molecule resolution and connects it to individual transcript isoforms. It uses approaches such as RiboSTAMP, in which an RNA-editing enzyme is fused to a ribosomal protein so that translating ribosomes leave a chemical mark on the RNA they read. Paired with long-read sequencing platforms such as Oxford Nanopore direct RNA sequencing, full-length transcripts and how they are translated can be read one molecule at a time.

Within ORFeus Work Package 1, DC2 is the transcript-level detection arm: it supplies high-resolution, isoform-level experimental data that grounds the computational predictions built across the network, and it connects to partner projects on AI-based smORF prediction, rare translation events, RNA modification effects on translation, and single-cell translation.

• Build multi-omics computational methods that combine RiboSTAMP data with other translation modalities (ribosome profiling (Ribo-seq), polysome profiling, etc) to map translation onto specific RNA isoforms. Quantify how features such as RNA structure and RNA modifications such as m6A influence translation, based on suitable RNA foundation models. Here we will draw on models, data and expertise from the projects of other doctoral candidates (DCs) in the network, particularly DC1 and DC7.
• Contribute to validating the ongoing implementation of RiboSTAMP in the lab, under normal conditions and perturbations (stress), to arrive at stable estimates of differential translation.
• Test whether RiboSTAMP can be run at single-cell resolution and generate isoform-level estimates of translation level and variance, for integration with DC1 AI predictions, DC3 rare-event catalogs, and DC9 single-cell validation.
• Collaborate across the ORFeus network, share validated protocols and isoform-level datasets via the ORFeome platform, and produce the project's scientific report.

Methods and platforms
: bioinformatics and genomics; machine learning and deep learning for genomics; RNA structure and modifications; single-molecule direct sequencing and ribosome profiling (Ribo-seq) data analysis; working with public and consortium datasets and models via the ORFeome platform and Hugging Face.

Secondment
: you will spend around three months at Oxford Nanopore Technologies (Oxford, UK), developing and optimizing nanopore-based direct RNA sequencing and analysis workflows for RiboSTAMP applications. You will also be guided by an independent academic advisor, with the possibility of a short, primarily virtual research exchange to strengthen the project.

MSCA eligibility

You must meet all of the following on your recruitment date:

• You do not already hold a doctoral degree. If you have defended a doctoral thesis but the degree has not yet been formally awarded, you are not eligible.
• Mobility rule: you must not have lived or carried out your main activity (work, studies, and so on) in Germany for more than 12 months in the 36 months immediately before your recruitment date. Compulsory national service, short stays such as holidays, and time spent in a procedure to obtain refugee status under the Geneva…