PHARMACELERA SL

Their core work

Pharmacelera is a Barcelona-based computational chemistry software company that builds AI-driven and quantum mechanics-based platforms to predict how drug candidates behave inside the body — specifically their absorption, distribution, metabolism, excretion, and toxicity (ADME-Tox) profiles. Their core product is a molecular modelling engine that replaces costly wet-lab experiments with fast computational screening, allowing pharmaceutical teams to eliminate poor candidates early in the preclinical pipeline. They apply machine learning methods — including one-shot learning for sparse data regimes — on top of physics-based molecular representations derived from quantum mechanics. In practice, they serve drug discovery teams that need to prioritize which compounds to synthesize and test next.

What they specialise in

How they've shifted over time

Pharmacelera began in 2018 with a narrower neural network approach to preclinical drug prediction (MolPredict — a small SME Phase 1 feasibility grant at €50k), focused on validating the concept that machine learning could boost preclinical research hit rates. By 2020 they had secured a full SME Phase 2 grant (€958k) for PharmScreen2, which shows a clear technical deepening: the modelling is now explicitly grounded in quantum mechanics, the ML layer has grown to include one-shot learning for sparse training data, and quantum computing appears as a forward-looking pillar. The trajectory is from proof-of-concept AI for drug screening toward a physics-grounded, quantum-ready platform — a significant step up in technical ambition and commercial scalability.

Pharmacelera is moving toward quantum-enhanced molecular simulation — organisations building consortia around quantum computing in life sciences or next-generation drug discovery platforms should watch them as a specialist software partner.

How they like to work

Pharmacelera has acted as coordinator on both of their H2020 projects, using the SME Instrument pathway — a programme designed for single companies driving their own commercial R&D, with no consortium partners required. This means their EU project track record shows no evidence of multi-partner collaboration under H2020; they work independently rather than as consortium nodes. A potential partner should expect them to bring a well-defined software product or toolset to a collaboration, rather than a history of shared project governance.

Pharmacelera's H2020 record shows zero registered consortium partners and zero countries collaborated with under the programme — a direct consequence of using the SME Instrument, which funds solo company projects. Any industrial or academic collaborations they have would appear outside the CORDIS record.

What sets them apart

Pharmacelera occupies a specific niche at the intersection of quantum mechanics-based molecular modelling and AI — a combination few Spanish SMEs can claim in the drug discovery software space. Their progression from a €50k feasibility study to a €958k Phase 2 platform grant within two years signals a company that successfully translated a research concept into a commercially viable tool. For pharma or biotech partners needing to accelerate early-stage compound screening without wet-lab costs, Pharmacelera offers a ready-to-deploy computational screening engine rather than a bespoke research service.

Highlights from their portfolio

• PharmScreen2
  The flagship project — at €958k it is one of the larger SME Phase 2 awards in computational chemistry, and it uniquely combines quantum mechanics-based modelling with one-shot machine learning and quantum computing, signalling a technically advanced platform play.
• MolPredict
  The seed project that de-risked the concept: a €50k SME Phase 1 grant in 2018 that validated neural-based prediction for drug preclinical research and directly unlocked the larger Phase 2 funding.