Background
I hold a PhD in Molecular Genetics from the University of Vienna (2004), and a diploma in the same subject completed with distinction (2001). My doctoral research was carried out at Intercell AG (now Valneva) and centred on machine-learning models for B-cell epitope prediction on pathogen proteins — which put me at the intersection of immunology and computation from the start.
Since then the work has been a mix of employed positions and independent consulting under the Sodatana umbrella. The recurring shape has been: a hard biological question, some data, a wall of computational possibilities — and the task of choosing well and moving the project forward. I value working in close contact with experimental scientists, and find the most satisfying projects are those where the computational and wet-lab sides genuinely inform each other.
Career
- Led an AI-driven protein engineering project focused on therapeutic enzyme discovery and activity/PK engineering, coordinating bioinformaticians, AI specialists and life scientists
- Hands-on work on chimeric lytic proteins (RiboLysins): engineering enzymatic and binding domains under the specific constraints of in vivo expression
- Built a data-driven candidate search pipeline drawing from scientific literature, clinical trial databases and news sources
- Served as liaison between computational/AI units and high-throughput wet-lab validation teams; set up an Airtable-based data platform for continuous experimental monitoring
- Independent bioinformatics consulting across infectious disease, oncology and diagnostics: vaccine design, protein and mRNA engineering, IP-aware R&D strategy
- Set up AWS cloud infrastructure for AI workloads (VPC using Terraform, EC2, EFS, ECR); de-novo protein ligand design using FreeBindCraft and BoltzGen
- Long-term collaboration with Sterogene Bioseparations on plasma purification; various peptide-based selections for diagnostic antibodies and vaccines
- Developed an algorithm for mRNA optimization; peptide fragment-based ligand design methods; drug repositioning studies using protein/drug/disease networks
Head of Vaccines (2012–2014) · Head of Therapeutic Innovation (2010–2011) · Senior Bioinformatician (2008–2010)
- Ran contract research projects end-to-end, typically delivering a vaccine design or analytical result within 2–4 weeks; projects commonly involved close client contact and covered infectious disease, oncology and veterinary pathogens
- Designed antigens for two synthetic veterinary vaccine prototypes that outperformed competitor products in animal model validations
- Built and operated a reverse-vaccinology pipeline for viral and bacterial pathogens; developed a comprehensive data integration platform (protein annotations, clinical trials, patents, biomedical literature) for biomarker identification
- Applied systems biology approaches (protein networks/pathways) for small-molecule drug repositioning in oncology and latent infections
- Immunoinformatics services (vaccine design), patent informatics and automated patent retrieval
- Internship at CeMM (Centre for Molecular Medicine): mass-spec data analysis and next target identification
- PhD thesis: machine-learning models for B-cell epitope prediction on pathogen proteins; co-contributor to the S. aureus antigen identification pipeline published in PNAS (2002)
- Invented a proprietary method for in-vitro epitope selection and evolution; co-inventor on S. aureus antigen patent (WO 2004/013166)
- Built automated pipelines for microbial sequencing data processing; developed a microbial genome viewer for interpreting complex genomic data
What I focus on now
The shift in the field toward AI-driven methods has been substantial, and I have moved with it while keeping classical biology and careful evaluation at the centre. Current work sits in three areas:
- AI-driven protein & de-novo design — hands-on with AlphaFold, RFdiffusion, ProteinMPNN, FreeBindCraft, BoltzGen and ligandMPNN, combined with classical structure-guided engineering. Cloud infrastructure (AWS VPC, EC2, EFS, ECR using Terraform) set up specifically for these workloads.
- Vaccine and mRNA design — a longstanding core: computational vaccinology, T-cell and B-cell epitope prediction, HLA and population coverage analysis, and tissue-specific mRNA optimization.
- AI/ML workflows and data infrastructure — LLM-based research pipelines (LangChain), bespoke scoring functions and predictors, and reproducible AWS environments for teams that need more than off-the-shelf tools.
How I tend to work
What I enjoy most is when a client leaves an engagement with more creative and technological capacity than they started with — not just a deliverable, but the ability to keep going.
The most valuable thing I can offer is honest scoping — including, when relevant, saying that a question is better answered by a wet-lab experiment than by another round of computation.
I work directly with clients (no agency layer, no offshore subcontracting). Projects range from a few days of focused work to multi-month embedded collaborations. IP stays with you; NDAs are routine and welcome. I find the most productive collaborations happen when there is genuine back-and-forth with experimental scientists — not just a computational output handed over a wall.
Publications & patents
Twenty-plus years of peer-reviewed work — fourteen papers and book chapters, one preprint under journal review, and three patent applications as co-inventor. The threads: B-cell epitope prediction and the machine learning around it, computational vaccinology as a reproducible workflow, systems biology for biomarkers and drug repositioning, and most recently engineered lytic proteins against S. aureus.
Selected
Badarau A, Bauer B, … Söllner J, … Corsini L. Multi-dimensional optimization of a lysin towards a ribolysin against life-threatening S. aureus infections: Fc-LysM-CHAP and its strong synergy with standard-of-care antibiotics. bioRxiv 2026; submitted to Nature Biotechnology. DOI
Söllner J. Computational peptide vaccinology. Methods Mol Biol. 2015;1268:291–312. PMID: 25555730
Söllner J, Heinzel A, Summer G, Fechete R, Stipkovits L, Szathmary S, Mayer B. Concept and application of a computational vaccinology workflow. Immunome Res. 2010;6 Suppl 2:S7. PMID: 21067549
Etz H, Minh DB, Henics T, Dryla A, Winkler B, Triska C, Boyd AP, Söllner J, et al. Identification of in vivo expressed vaccine candidate antigens from Staphylococcus aureus. Proc Natl Acad Sci USA. 2002;99(10):6573–8. PMID: 11997460
Full publication & patent record →
Working together
If any of the above resonates with what you're trying to do, the simplest next step is a short email outlining the question. I'll come back honestly about whether I'm the right fit and what a sensible first engagement might look like.