DLAI

Designing Life with AI

EPFL MAKE — Protein Design Association

bringing ai protein
design to students

MAKE Association · EPFL

DLAI is a student association at EPFL dedicated to designing proteins with AI and computational tools, then testing them in the lab to see if the engineering really works.

We create new binders, metal‑binding proteins and immune‑targeted designs in collaboration with research labs across campus, turning cutting‑edge methods in protein engineering into hands‑on projects for students.

Our mission

We aim to bridge academia and student experience by connecting undergraduates and graduate students with EPFL labs that support our projects, offer guidance and propose real design challenges.

Through workshops, design sprints and MAKE projects, we explore how AI‑based models and physical experiments can work together to build useful, creative and responsible new proteins.

the people
behind dlai

We are engineering, life‑science and computer‑science students with a shared interest in protein design, synthetic biology and AI.

designing, building
and testing proteins

Our projects span AI‑driven sequence design, structure prediction, and experimental validation in collaboration with EPFL labs. Here you can keep track of ongoing work and also calls for open projects.

Phosphosite engineering protein In progress

Phosphosite engineering

This project combines molecular biology and artificial intelligence to investigate conformational changes in phosphosites upon phosphorylation, analyzing over 20,000 phosphosites using custom Python scripts and refined selections with Alphafold predictions. The next phase involves validating about 100 selected phosphosites using NMR spectroscopy to confirm AI model predictions, potentially advancing our understanding of protein functionality and leading to groundbreaking applications in pharmaceuticals and biomolecular studies.

Light-activated kinases

Light activated kinases

We are trying to modify protein kinases by removing some of their domains but such that they keep their usual functionality in cells. Once we have these we'd like to fuse them to other proteins / protein domains to be able to better regulate the protein's activity (eg. addition of a light activating domain so that light would act as an on/off switch).

Hydrophobic mutants of proteins

Hydrophobic mutants of proteins

The goal was to find hydrophobic mutants of proteins while keeping the shape by randomized mutations. The hydrophilic amino acids on the surface have been replaced by hydrophobic ones.

EphA2 receptor binders

Designing and testing EphA2 receptor binders

The goal is to design and test binders for EphA2 receptors, which are receptors overexpressed in tumoral cells.

Alcoholic fermentation pathway

An alternative alcoholic fermentation pathway in yeast

We are four life sciences engineering students working on optimizing ethanol production through yeast anaerobic fermentation. Our goal is to establish an alternative fermentation pathway in yeasts by studying the integration of a bacterial fermentation pathway presenting advantages such as alleviating anaerobic redox stress or bypassing toxic intermediates. The efficiency of the enzymes involved will then be optimised using novel AI tools such as RFdiffusion.

URE3 prion binder

[URE3] prion binder

We want to create a protein binder that binds to the [URE3] prion in S. cerevisiae, with the future goal of being able to design a binder for prions causing diseases in humans.

Probe-free imaging method

Probe‑free non‑invasive imaging of intracellular antigens

This project aims at developing a novel non‑invasive, probe‑free method for imaging intracellular antigens.

GPCR biosensor

GPCR biosensor development

Advancing molecular detection via BRET and synthetic biology. The GPCR Biosensor project focuses on detecting the activation of Gαi/t‑subunits in yeast through GPCRs using a BRET (Bioluminescence Resonance Energy Transfer) system. Our current method involves the KB1753 peptide, acting as a proof of concept in laboratory conditions. The aim is to develop this system to recognize a range of molecules, like histamine, by modifying the GPCR binding site. This could lead to significant advancements in understanding signal transduction and drug interactions. The team's efforts are grounded in seminal research, with foundational work by C. Johnston et al. in Biochemistry, 2006, and R. Jefferson et al. in Nature, 2023, focusing on the minimal determinants for Gαi/t binding and the dynamic receptor‑peptide signaling complexes, respectively. This project represents a convergence of molecular biology, biochemistry, and synthetic biology, aiming to create a versatile tool for molecular detection and study.

partner labs
& supporters

DLAI works closely with EPFL research groups and external partners who provide scientific guidance, lab space and design challenges. Here is an overview of our hosting laboratories.

To see all open projects, go to the open projects page.

Sahand Rahi lab

Prof Rahi's lab focuses on Timing of Cellular Decisions, Directed Evolution, AI- and Physics-based Design of Computational Protein Functionalities, and Neural Computation Across Timescales.

Dal Peraro — Molecular Biomodelling Lab

The main goal of Dal Peraro's laboratory is to understand the physical and chemical properties of complex biological systems, in particular their function emerging from structure. To address these questions, they use and develop a broad spectrum of computational tools fully integrated with experimental data.

Correia — Protein Design and Immunoengineering lab

Correia's group is driven by the passion of expanding nature's repertoire by computationally designing novel functional proteins to be used for practical purposes such as therapeutics, vaccines, biosensors and others.

Barth — Laboratory of Protein and Cell Engineering

The LPCE works at the interface of biophysics, chemical, structural, computational and cell biology to uncover the molecular principles that regulate protein and cellular signaling, designing protein systems with novel biosensing and signaling functions and predicting the effects of genetic variations for personalized cancer medicine.

De Los Rios — Laboratory of Statistical Biophysics

Established in October 2003 on the EPFL/UNIL campus, the lab uses tools mainly from statistical physics to contribute to the understanding of the properties of living matter, working across statistical physics, biophysics and computational physics.

Supported by

stay
connected

Follow our activities, events and project updates on social media and EPFL channels.

Before you join

Please carefully read the expectations for newcomers.

If you are interested in our project, join our WhatsApp channel for updates.

General contact

For any supplementary information on this MAKE project, please contact one of the following:

About a specific project

Write an email to dlai@epfl.ch and cc one person from the list. You can find project names, labs and info on the open projects page.