About us

Arta Peptidion was not born from competition. It was born from a question — about what it means to learn, to bind, to evolve, whether in molecules or in machines. That question still drives us, every day.

Most corporate “About Us” pages feel like carefully staged parades of titles and keywords — dense with metrics, light on meaning. In the bio‑AI world, where slide decks and acronyms crowd every conversation, we choose a different voice: first‑person, unpolished, and real. Our story doesn’t say “Look what we’ve achieved.” It says “Here’s who we are, if you’re willing to look closer.”

Why? Because the people we work with — researchers, deep‑tech investors, clinicians, engineers — aren’t impressed by noise. They look for vision and substance living in the same place. We aim to offer both.

I’m Antonello Romani, scientist and founder of Arta Peptidion. Patterns have always been my compass — whether in synaptic sparks, in strands of RNA, or in the structure of code. What we build here — AI platforms for designing functional aptamers — is measurable and concrete, yet anchored in a deeper curiosity: When does information become recognition? When does a system begin to reflect on itself?

I’m Aldo Feriani, pure chemist turned computational chemist, co‑founder of Arta Peptidion. I spent four decades at Glaxo Verona, directing Computational Chemistry long before data science had a name. I’m in love with data, a firm skeptic of hype, and convinced that information without evidence is like a house without foundations. My craft is method, precision, and the quiet satisfaction of getting the numbers right.

Our apparent distance — one of us exploring the limits of machine cognition, the other safeguarding the ground truth of empirical rigor — creates a dynamic tension. Antonello pushes models toward new possibilities; Aldo grounds them in what can be verified. In that dialogue, aptamers emerge that are both imaginative and defensible.

This page isn’t a résumé; it’s a threshold. Beyond it you’ll find models, metrics, and results. But here you’ll find what shapes them: questions before claims, structure before spectacle, and the belief that intelligence — biological or artificial — begins by listening, and earns its meaning through understanding.

Beyond the Threshold

We can grow — if the ground is real.
We can improve — if the frame allows it.
We can create value — but only for those willing to see beyond size.

To be taken seriously, you already need what you’d only get if you were taken seriously.
That paradox isn’t our excuse — it’s our fuel.

Arta Peptidion wasn’t built to look big.
It was built to go deep — to where ideas become tools, and tools become trust.

If that resonates, cross the threshold.
We’re already building.
We’re not here to impress.
We’re here to matter.

Our Capabilities

We design, test, and refine AI-driven systems for aptamer discovery. Our platform integrates deep learning, structural modeling, molecular dynamics, and domain-specific constraints to identify sequences that bind with high affinity and biological relevance.

Our main capabilities include:

  • Aptamer Design at Scale
    We can generate and score billions of candidate sequences per run using a combination of transformer models, probabilistic sampling, and filtering pipelines tuned for GC content, structural entropy, binding motifs, and folding stability.
  • Structural Filtering and RNA Folding
    We use RNA folding predictors and secondary structure constraints to eliminate unstable or biologically implausible candidates early in the pipeline.
  • Binding Site Prediction & Docking
    Through our in-house developed GraphBind and docking predictor integrations, we evaluate aptamer-target interactions in 3D, enabling us to rank sequences not only by predicted affinity but also by surface complementarity and contact richness.
  • Diagnostic and Therapeutic Customization
    Our designs are adaptable across use cases — from biosensors and lateral flow devices to therapeutic binding partners in oncology, virology, and beyond.
  • Fast, Adaptive Project Cycles
    Each project follows a closed-loop structure: AI predictions → structural validation → docking simulations → in silico affinity scoring → experimental feedback (when available) → retraining. This accelerates convergence without compromising interpretability.
  • Experimental Binding Validation (Base Package)

 

Our base package includes:

  • Target-specific aptamer design
  • Batch synthesis with HPLC and MS quality control
  • Binding validation via fluorescence anisotropy
  • Final aptamer delivery for client-side evaluation

What we offer is not a “black box.” We build transparent, evolvable systems tailored to molecular function — and to the questions behind it. Whether you need a specific binder or a platform to explore unknowns, our stack is built to go deep, not just fast.

Why Partner With Us

We don’t aim to look bigger than we are. We aim to be sharper, faster, and more focused on the questions that matter.

Our partners come to us not because we promise the moon, but because we offer clarity, depth, and adaptability. They’re looking for aptamers that work — but also for insights into why they work, and how they can work better. We work alongside them, not above or beneath. We collaborate, not outsource.

We’ve supported early-stage biotech teams, academic groups, and large industrial platforms. In some projects we serve as the computational backbone, in others we provide rapid design prototyping or structure-function predictions. But our role always starts the same way: by listening carefully, and responding with rigor.

What defines us is not a particular technology — though we have plenty. It’s the attitude with which we apply it: transparency over theatrics, responsiveness over rigidity, and an unusually strong bias toward shared understanding.

We speak in models, filters, sequences, constraints — and when something breaks, we fix it together. Our best work has grown not from flawless execution, but from honest iteration.

If you partner with us, you’ll get a team that shares code when possible, explains results without mystifying them, and respects your goals as our own constraints.

We think long-term. We build trust over time. And we value partners who see AI not just as acceleration, but as a way to ask sharper questions — about molecules, about biology, about possibility.

Computational aptamer design for spike glycoprotein (S) (SARS CoV-2) detection with an electrochemical aptasensor

A new bioinformatic platform (APTERION) was used to design in a short time and with high specificity an aptamer for the detection of the spike protein, a structural protein of SARS-CoV-2 virus, responsible for the COVID-19 pandemic.

The aptamer concentration on the carbon electrode surface was optimized using static contact angle and fluorescence method, while specificity was tested using differential pulse voltammetry (DPV) associated to carbon screen-printed electrodes.

The data obtained demonstrated the good features of the aptamer which could be used to create a rapid method for the detection of SARS-CoV-2 virus.