Michael T. Wolfinger has taught computational RNA biology since 2005, in university courses at the University of Vienna, the University of Freiburg and FH Campus Wien, and in invited workshops for research teams. The formats below grow out of that teaching. They are practical: participants work on real sequences and leave with workflows they can run again. Content and pacing are adjusted to the group before the session.

ViennaRNA and computational RNA design

In preparation

This material has been taught in university courses and small groups. As an RNA Forecast training it is being prepared, and dates will be announced to the waitlist first.

Who it is for. Experimental RNA labs that want computational depth, doctoral programmes, biotech teams working on mRNA, siRNA or structured RNA, computational groups and tool developers.

What is covered, chosen per group, half day to full day:

RNA secondary structure and the nearest neighbour model: what RNAfold computes, and why the MFE structure is one answer among many.
The Boltzmann ensemble, i.e. pair probabilities, centroid and MEA structures, ensemble diversity and helix probabilities, how to read a dot plot, and when the MFE misleads.
Probing data, i.e. SHAPE and DMS reactivities as soft constraints, the three ViennaRNA methods and, new in ViennaRNA 2.8.0, cross-linking interval pairs, with a look at what the data change and what they do not.
Modified nucleotides, i.e. the pseudouridine and N1-methylpseudouridine parameters for therapeutic mRNA, so that the molecule that is folded is the one that is made.
Large RNAs and viral genomes: local folding with RNAplfold, locally stable elements, conserved structure from alignments with RNAalifold, covariation as evidence.
Accessibility and RNA–RNA interaction: RNAplfold, RNAcofold and RNAup for siRNA, ASO and miRNA sites, and how to map variants onto a structure.
Design review before synthesis, i.e. the objective, the ensemble defect and how a design holds up under sequence variants, with a checklist that participants take home.

Tertiary structure prediction is outside the scope. The course covers how far secondary structure and ensembles take a question before that becomes necessary.

Format. Live, remote or on site, as a group of 6 to 12 or as a single participant in an open remote cohort, half day or full day, participants' own sequences welcome. Materials stay with the group.

Free material on the topic, in reading order on the ViennaRNA in practice page: RNAfold and the ensemble, SHAPE and chemical probing data, RNAplfold and accessibility, the ViennaRNA 2.8.0 release note and the RNA structure glossary.

Waitlist

Something was missing or not valid. Please check the fields and send again.

Name

Email *

Affiliation

You are

Preferred format

Who would attend

What should it cover?

You get one e-mail when a date is fixed and nothing else, and a reply to that e-mail takes you off the list.

Problems with the form? Email hello@rnaforecast.com.

Local AI coding tools for research groups and small teams

In preparation

The setup guides this course is built on are published and maintained on michaelwolfinger.com. The course itself is in preparation and will run as a live half-day cohort, updated for every run because the tools change monthly.

A scientist's workshop on running coding agents, Claude Code, OpenAI Codex CLI, OpenCode and pi, against models that run on your own hardware with LM Studio, Ollama or an exo cluster across several machines, on Apple Silicon and on Linux with NVIDIA or AMD GPUs, with notes for Windows. For research groups and small teams whose data must stay in house: unpublished data, manuscripts and code that must not leave the machine.

What is covered:

Hardware and model choice, i.e. what fits in 16, 24, 32 and 64+ GB with measured memory needs, tool calling as the property that decides whether a model is usable at all (and how to check it instead of trusting the agent's report), and context size as the second limit.
Setup on macOS and Linux, with the differences on Windows, i.e. LM Studio, Ollama, the environment variables and the context length, which is the setting that breaks most sessions.
A research workflow: scripts and analysis notebooks, tests as the safety net, AGENTS.md and CLAUDE.md for durable instructions, reading diffs.
What local models can and cannot do compared with hosted ones, where they fail, and how to work around it.
Privacy and cost: what leaves the machine with each configuration, telemetry switches, what a GPU box costs against API spend.
Hands-on: every participant leaves with a working setup on their laptop or against the group's server.

Prerequisites. A laptop with 16 GB or more, or access to a group machine, and some comfort with a terminal. Format. Live cohort, half day, 6 to 12 participants, remote or on site.

The guides: Claude Code with a local model on Apple Silicon, Claude Code with Ollama, Codex CLI with LM Studio, Ollama or exo, four coding agents against ten local models and a reproducible data analysis with a local model.