Summer School 2026

Machine Learning in Biology

A four-day summer school equipping life scientists with the practical skills to apply machine learning across real biological datasets.

15–18 June 2026 · Mon–Thu University of Bern · Main Building, Room 304 30 seats · first-come, first-served

About the Course

This four-day summer school equips life scientists with the practical skills to apply machine learning across real biological datasets. Mornings cover the methodological foundations — supervised and unsupervised learning, model evaluation, and the pitfalls specific to biological data. Afternoons are spent in hands-on coding sessions working with curated datasets — the Palmer Penguins, synthetic gene-expression data, and a microbiome dataset.

The programme includes a one-day mini-symposium on Wednesday in which researchers present current applications of machine learning in biological and biomedical research, providing context for how the methods covered translate into ongoing science.

Who should attend

PhD students, postdoctoral researchers, and scientific staff in the life sciences who want to move beyond using machine learning as a black box.

Prior experience with R is helpful but not required; participants without programming background should expect to invest additional effort during the practical sessions.

What You Will Gain

A Working Vocabulary

The ability to read ML methods sections critically and discuss them with collaborators.

Hands-on Competence

You will train, validate, and interpret models on real biological datasets — from the Palmer Penguins to gene-expression and microbiome data — during the course.

Judgement

Recognising when ML is the right tool, which method fits the question, and how to avoid the most common failure modes — data leakage, overfitting, batch effects.

A Network

Four days of close interaction with instructors, invited speakers, and peers from across Swiss life-science institutions.

Programme

Monday
15 June
Foundations & Regularization
09:30
Introduction & Welcome
10:00
Coffee Break
10:30–12:00
Linear & logistic regression
From questions to baseline models; over- vs. underfitting
12:00–13:00
Lunch
13:00–15:00
Regularization: ridge & lasso
Cross-validation & model selection
15:00
Coffee Break
15:30–17:00
Practical
Calibration intuition & data cards
Datasets Synthetic gene-expression data · Palmer Penguins
Tuesday
16 June
Trees & the tidymodels Pipeline
08:30–10:00
Decision trees
The “two cultures” of modelling
10:00
Coffee Break
10:30–12:00
tidymodels pipeline in R
recipe → workflow → tune → fit
12:00–13:00
Lunch
13:00–15:00
Hyperparameter tuning
Reproducible project layout
15:00
Coffee Break
15:30–17:00
Practical & Discussion
Microbiome lab
Datasets Synthetic genes · Palmer Penguins · Microbiome (OTU)
Wednesday
17 June · Mini-Symposium
10:00
Coffee & Welcome
10:30–12:00
Invited Talks
Guillaume Witz — MIC, University of Bern
Robin Zbinden — EPFL
12:00–13:30
Lunch
13:30–15:30
Invited Talks & Discussion
Cyril Malbranke — EPFL
Marco Baity-Jesi — Eawag
Thursday
18 June
Advanced Models & Interpretability
08:30–10:00
Forests, boosting & neural nets
The model catalog
10:00
Coffee Break
10:30–12:00
PCA, imputation & resampling
Preprocessing as recipe steps
12:00–13:00
Lunch
13:00–15:00
Metrics & class imbalance
Confusion matrices, ROC / PR
15:00
Coffee Break
15:30–17:00
Variable importance & SHAP
Interpretation guardrails; closing
Datasets Palmer Penguins (subgroup checks) · Microbiome
Note: Taught in R (RStudio or Positron) using the tidymodels ecosystem. Two running data tracks — the Palmer Penguins (real measurements) and synthetic gene-expression data (known ground truth) — are reused across the week, with a microbiome dataset in the Tuesday and Thursday labs. Wednesday's symposium speakers are confirmed; the detailed timetable will be finalised here. Course materials — slides, notebooks, and lab exercises — are available on the course site →

Faculty & Speakers

Lecturers

  • Dr. Stephan Peischl
  • Dr. David Miguel Ferreira Francisco
  • Dr. Aparna Pandey
Interfaculty Bioinformatics Unit (IBU) · University of Bern

Invited Speakers

  • Robin ZbindenPhD candidate · ECEO, EPFL
  • Dr. Guillaume WitzMIC & Data Science Lab, University of Bern
  • Dr. Marco Baity-JesiML & Complex Systems Group, Eawag
  • Cyril MalbrankeEPFL
Wednesday symposium · 17 June

Registration & Costs

Participation is funded by a cost contribution per participant — this is not a registration fee. CUSO PhD students pay no contribution; their eligible costs are reimbursed directly via CUSO.

CUSO PhD Student
CHF 0
cost contribution
CHF 180 with hotel (3 nights) — or book your own
No cost contribution. Course materials, lunches, coffee breaks, and the social event are covered.
CUSO reimburses eligible participation costs directly via myCUSO; the organised hotel carries a CHF 180 personal share. Travel reimbursed separately.
Non-CUSO Participant
CHF 190
cost contribution
CHF 670 with hotel (3 nights)
Includes course materials, lunches, coffee breaks, and social event.
Hotel: 3 nights incl. breakfast near the University of Bern.
Register Now
Registration opens 11 May 2026 · Deadline 8 June 2026 · 30 seats, first-come first-served

Venue

The summer school takes place in Room 304 of the Main Building (Hauptgebäude) of the University of Bern, centrally located in the heart of the city.

Bern is easily reachable by train from all major Swiss cities. The main building is a 10-minute walk from Bern central station.

Directions and building information →

University of Bern — Main Building

Hochschulstrasse 4

3012 Bern, Switzerland

Room 304 (3rd floor)