☰

Sibylla Biotech S.p.A.

Integrated data for an IND filing

A computational drug discovery company scaling its PPI-FIT platform toward an IND filing

Industry
Biotechnology (Computational Drug Discovery)
Headquarters
Bresso, Milan, Italy
Public information as of
January 2026

A4BEE prepared this analysis from publicly available sources. It reflects our own reading of Sibylla Biotech S.p.A.'s published strategy and is not endorsed by, or produced in cooperation with, Sibylla Biotech S.p.A.. Company website

Strategic priorities

Sibylla Biotech S.p.A. is an Italian biotech company commercialising the PPI-FIT computational platform for targeted protein degradation drug discovery. The platform identifies 'undruggable' protein pocket combinations to develop Cyclin D1 degrader candidates for oncology and CNS disorders, with strategic partnerships with Takeda, Ono Pharmaceutical and MD Anderson Cancer Center validating its commercial potential. The company's immediate priority is transitioning from research-centric operations to GxP-compliant clinical infrastructure ahead of IND filings, requiring fundamental digital infrastructure development alongside the scientific programme.

The defining challenge is the wet-to-dry lab synchronization gap. Sibylla's computational simulations and biological validation experiments operate at physically separated Bresso facilities, creating latency in the experimental feedback loop that is critical for hit-to-lead optimisation. Medicinal chemists may optimise compounds based on outdated biological data when assay results remain trapped in local workstations rather than integrated with simulation models. The HPC infrastructure — required for all-atom protein folding simulations for complex oncological targets including KRAS — is constrained by localized workstation capabilities that cannot scale to industrial production levels. Any interruption in computing power delays hit-to-lead cycles by months against well-funded AI drug discovery competitors.

On the data side, the Biology Lab, MedChem Lab and Computational Lab operate with fragmented LIMS systems and Excel Islands that prevent unified data visibility across the drug discovery pipeline. Scientists spend significant time on manual data cleaning and reconciliation rather than discovery science. As the Cyclin D1 programme advances toward IND filing, laboratory workflows must meet FDA/EMA data integrity standards (ALCOA+ principles) that academic-origin processes typically lack, and partner data exchange with Takeda, Ono and MD Anderson requires compliant data sharing infrastructure that does not yet exist.

Challenges we see

  • Operations Integration

    Slowing experimental feedback loop as assay data remains trapped in local workstations

    Computational simulations and biological validation experiments operate at physically separated Bresso facilities. Assay results remain trapped in local workstations rather than integrating with simulation models, creating latency in the experimental feedback loop that is critical for hit-to-lead optimisation.

    Where wet-lab data is not connected to the computational model, medicinal chemists optimise compounds against stale biological assumptions. A unified data platform means the simulation model is always calibrated to the most recent assay data.

  • Digital Manufacturing

    HPC workstation constraints limiting protein folding simulation for complex oncology targets

    All-atom protein folding simulations for KRAS and Cyclin D1 require computational resources that exceed current localized workstation capabilities. Any interruption in computing power or data storage delays hit-to-lead cycles by months, creating competitive disadvantage against better-resourced AI drug discovery rivals.

    Where HPC capacity is constrained by localised infrastructure, the simulation throughput is set by the workstation rather than by the scientific programme. Cloud-based HPC infrastructure means the computational capacity scales with the programme, not with the hardware budget.

  • Digital Integration

    Fragmented LIMS and Excel Islands preventing unified drug discovery data visibility

    The Biology Lab, MedChem Lab and Computational Lab operate with fragmented LIMS systems and Excel Islands that prevent unified data visibility across the drug discovery pipeline. Scientists spend significant time on manual data cleaning and reconciliation rather than discovery science.

    Where data is fragmented by lab function, the time spent reconciling spreadsheets is time not spent on discovery. A unified data environment means scientists have the data they need in the format they need it, without manual reconciliation.

  • Compliance Regulatory

    Manual workflows creating ALCOA+ compliance gaps ahead of IND filing

    As the Cyclin D1 programme advances toward IND filing, laboratory workflows must meet FDA/EMA data integrity standards (ALCOA+ principles) that academic-origin processes typically lack. Manual workflows without automated audit trails create risk of data misattribution, inconsistency or loss.

    Where compliance depends on manual procedures rather than system-enforced controls, the audit readiness is assembled retrospectively rather than maintained continuously. A GxP-compliant data platform means ALCOA+ compliance is built into the system, not documented in a folder.

  • Operations Integration

    Partner data exchange infrastructure not yet in place for Takeda, Ono and MD Anderson

    Strategic partnerships with Takeda, Ono Pharmaceutical and MD Anderson Cancer Center require compliant data sharing infrastructure that does not yet exist. Partner data exchange is currently managed through informal channels without systematic data quality controls or audit trails.

    Where partner data exchange is informal, the data quality and audit trail requirements of pharmaceutical partnerships are not met. A compliant data exchange platform means partner data flows are governed, auditable and ready for regulatory inspection.

Opportunities, by urgency and business impact

Each bubble is one opportunity, numbered to match the list below. Further right means it bites sooner; higher means a bigger effect on the business. A bigger bubble means a bigger implementation effort.

Source: A4BEE analysis of public sources

  1. Unified wet-to-dry lab integration platform for PPI-FIT

    Computational simulations and biological validation operate at separated Bresso facilities with assay data trapped in local workstations. The experimental feedback loop is delayed by the time it takes for data to physically move between buildings, giving medicinal chemists stale data to work with.

    Deploy a unified data platform that ingests assay results automatically from Biology Lab instruments and delivers them to the PPI-FIT computational environment in real time, closing the wet-to-dry lab feedback loop to hours rather than days.

    • Sibylla Biotech strategic account analysis, 2025

  2. Cloud-based HPC infrastructure for industrial-scale protein folding simulations

    All-atom protein folding simulations for KRAS and Cyclin D1 exceed the capacity of localized workstations, delaying hit-to-lead cycles by months and creating competitive disadvantage against better-resourced AI drug discovery rivals.

    Implement cloud-based HPC infrastructure that scales computational capacity with the programme requirements, enabling all-atom simulations for complex oncology targets without hardware procurement constraints.

    • Sibylla Biotech HPC requirements assessment, 2025

  3. GxP-compliant ALCOA+ data platform ahead of IND filing

    Academic-origin laboratory workflows lack the automated audit trails and data integrity controls required for FDA/EMA IND submissions. Manual data handling creates misattribution and inconsistency risk that could delay or derail clinical trial applications.

    Deploy a GxP-compliant data platform with ALCOA+ principles built in — electronic signatures, audit trails, data versioning — ensuring the clinical data package is submission-ready when the IND filing is prepared.

    • Sibylla Biotech regulatory readiness assessment, 2025

  4. Compliant partner data exchange infrastructure for pharma partnerships

    Data exchange with Takeda, Ono Pharmaceutical and MD Anderson Cancer Center is currently informal and unstructured, creating compliance risk and limiting the velocity of collaborative research activities.

    Build compliant data exchange infrastructure that enables governed, auditable data sharing with pharmaceutical partners, satisfying 21 CFR Part 11 requirements and accelerating collaborative research velocity.

    • Sibylla Biotech partnership infrastructure assessment, 2025

  5. AI-driven hit identification acceleration using integrated discovery data

    Fragmented LIMS and Excel Islands mean scientists cannot apply AI/ML to the full drug discovery dataset. The signal in the data is not fully exploited because the data is not in a single environment.

    Establish an integrated data environment that enables AI-driven hit identification across the full PPI-FIT dataset — computational predictions, assay results, MedChem SAR — accelerating the identification of degrader candidates for the Cyclin D1 and KRAS programmes.

    • Sibylla Biotech AI readiness assessment, 2025

What we'd propose

  • Digital Lab

    Unified wet-to-dry lab integration platform for Sibylla operations

    We design and deploy a unified data platform for Sibylla that integrates the Biology Lab, MedChem Lab and Computational Lab at Bresso into a single data environment, automatically ingesting assay results and delivering them to the PPI-FIT computational environment in real time.

    • Automated assay data ingestion

      Biology Lab instruments connected to the platform automatically

      Build automated data ingestion from Biology Lab instruments — plate readers, flow cytometers, assay plates — that captures results at the point of generation and routes them to the PPI-FIT computational environment without manual export steps.

    • Computational environment integration

      PPI-FIT always running on the latest assay data

      Implement integration between the assay data platform and the PPI-FIT computational environment, ensuring that simulation models are always calibrated to the most recent experimental data and that new computational predictions are immediately available to medicinal chemists.

    • Collaborative SAR workspace

      One workspace for the whole discovery team

      Deliver a shared structure-activity relationship workspace that gives the entire discovery team — computational chemists, medicinal chemists, biologists — simultaneous access to integrated assay and simulation data.

    • Experimental feedback loop closed from days to hours by eliminating manual data movement between facilities.
    • Computational predictions calibrated to the latest assay data, improving the quality of lead optimisation decisions.
    • Discovery team collaboration enabled by a shared data environment that spans all three labs.

    We've done this before

    See how we do Digital Lab
  • Enterprise AI

    Cloud-based HPC infrastructure for industrial-scale simulations

    We implement cloud-based HPC infrastructure for Sibylla that provides elastic computational capacity for all-atom protein folding simulations, enabling industrial-scale throughput for KRAS, Cyclin D1 and new oncology targets without hardware procurement constraints.

    • Cloud HPC cluster provisioning

      Computational capacity that scales with the programme

      Deploy cloud-based HPC clusters — AWS Batch, Google Cloud Life Sciences or Azure CycleCloud — with elastic scaling that provisions computational capacity on demand as simulation workloads fluctuate.

    • PPI-FIT workflow orchestration

      Complex simulation workflows automated end-to-end

      Build workflow orchestration for PPI-FIT simulation pipelines that automates the full sequence from target selection through all-atom folding to result analysis, with automated failure recovery and checkpoint restart.

    • Computational result management and sharing

      Simulation results stored, searchable and partner-accessible

      Implement a computational results management platform that stores, indexes and makes searchable all simulation outputs, enabling computational biologists and medicinal chemists to retrieve and compare historical simulation results.

    • Hit-to-lead cycle time reduced by removing HPC capacity as a constraint on simulation throughput.
    • Competitive position strengthened against better-resourced AI drug discovery rivals by accessing cloud-scale computational resources.
    • Capital expenditure on hardware replaced by operational expenditure that scales with actual usage.

    We've done this before

    See how we do Enterprise AI
  • Digital Lab

    GxP-compliant ALCOA+ data platform for IND-enabling studies

    We deploy a GxP-compliant data platform for Sibylla's clinical development programme that implements ALCOA+ principles — attributable, legible, contemporaneous, original, accurate — with electronic signatures, complete audit trails and 21 CFR Part 11 compliance for FDA/EMA IND submissions.

    • GxP-compliant LIMS deployment

      Laboratory data managed to ALCOA+ standards from day one

      Deploy a GxP-compliant LIMS that manages all laboratory data — sample tracking, assay results, reagent inventory — to ALCOA+ standards with electronic signatures, complete audit trails and automated data versioning.

    • Electronic laboratory notebook integration

      Research data captured contemporaneously in structured format

      Integrate an electronic laboratory notebook that captures experimental observations contemporaneously with structured data fields, eliminating the paper notebook compliance gap that IND reviewers frequently cite.

    • IND submission data package preparation

      Submission-ready data package assembled automatically

      Build automated IND submission data package preparation that assembles the complete clinical data package — raw data, analysis datasets, audit trails, metadata — in a format ready for FDA/EMA submission.

    • IND filing risk reduced by ensuring data integrity controls are system-enforced rather than procedure-dependent.
    • Regulatory inspection readiness improved through continuous compliance rather than retrospective documentation assembly.
    • Scientific team time redirected from compliance documentation to discovery science.

    We've done this before

    See how we do Digital Lab
  • Enterprise AI

    Compliant partner data exchange platform for pharma collaborations

    We build compliant data exchange infrastructure for Sibylla's pharmaceutical partnerships with Takeda, Ono Pharmaceutical and MD Anderson Cancer Center, enabling governed, auditable and 21 CFR Part 11-compliant data sharing that accelerates collaborative research.

    • Governed data sharing framework

      Partner data flows governed by agreed data quality standards

      Design and implement a governed data sharing framework that defines data quality standards, exchange formats and usage rights for each partnership, ensuring both parties have clarity on data handling obligations.

    • Secure data transfer infrastructure

      Data exchanged securely with complete audit trails

      Deploy secure data transfer infrastructure with encryption in transit and at rest, access controls and complete audit trails for all partner data exchange activities, satisfying 21 CFR Part 11 requirements.

    • Partner collaboration workspace

      Shared analytical workspace for each partnership

      Implement partner-specific collaboration workspaces that provide Takeda, Ono and MD Anderson researchers with access to relevant datasets and simulation outputs within a governed, audited environment.

    • Collaborative research velocity improved by removing informal data exchange as a bottleneck.
    • 21 CFR Part 11 compliance demonstrable to pharmaceutical partners through structured governance.
    • Partner relationship value deepened by providing compliant, auditable data access.

    We've done this before

    See how we do Enterprise AI
  • Enterprise AI

    AI-driven hit identification acceleration across the PPI-FIT integrated dataset

    We establish an AI-driven hit identification capability for Sibylla that applies machine learning across the unified PPI-FIT dataset — computational predictions, assay results, MedChem SAR — to accelerate degrader candidate identification for the Cyclin D1 and KRAS programmes.

    • Integrated discovery data warehouse

      All discovery data in one ML-ready environment

      Build an integrated data warehouse that consolidates computational predictions, assay results and medicinal chemistry SAR data into an ML-ready environment where algorithms can identify patterns across the full discovery dataset.

    • ML-driven hit identification models

      AI models trained on the full PPI-FIT dataset

      Develop ML models trained on the integrated PPI-FIT dataset to predict degrader activity from structural and computational features, prioritising candidates for experimental testing and accelerating the hit identification phase.

    • SAR knowledge graph

      Relationship knowledge captured and queryable

      Build a structure-activity relationship knowledge graph that captures and encodes the relationships between compound structures, computational features and biological activity across all historical experiments.

    • Hit identification accelerated by ML models that exploit patterns across the full discovery dataset.
    • Experimental resource allocation improved by ML-driven prioritisation of candidates for assay testing.
    • Institutional knowledge codified in the SAR knowledge graph rather than held by individual scientists.

    We've done this before

    See how we do Enterprise AI

Digital maturity: today and target

Scored out of 100 across six dimensions. The target is what Sibylla Biotech S.p.A.'s own published ambition implies — not a perfect score.

Source: A4BEE analysis of public sources
Wet-Dry Lab Integration 25 → 85
Computational simulations and biological validation operate at separated Bresso facilities with no automated data flow. Assay data requires manual export and transfer to the PPI-FIT computational environment.
HPC Infrastructure 30 → 90
Protein folding simulations run on localised workstations with constrained capacity. No cloud-based HPC infrastructure or elastic computational capacity is deployed.
Data Integration 25 → 80
Biology Lab, MedChem Lab and Computational Lab have fragmented LIMS and Excel Islands. No unified data platform spans the three labs.
GxP Compliance 25 → 90
Laboratory workflows are academic-origin with manual procedures. No GxP-compliant LIMS, electronic notebook or ALCOA+ data platform is in production for the clinical development programme.
Partner Data Exchange 25 → 75
Data exchange with Takeda, Ono Pharmaceutical and MD Anderson is informal and unstructured. No compliant data exchange platform or governed sharing framework is in production.
AI/ML Capability 20 → 70
AI/ML is not systematically applied to the drug discovery dataset. Fragmented data environments prevent the integrated dataset that ML-driven hit identification requires.

Check this yourself

Our Service Portal has free self-assessments and market comparisons. These are the ones that line up with what we've read above — no sales call required.

Think we've read this right?

Talk to us

Related reading

This is an independent analysis prepared by A4BEE from publicly available information as of January 2026. It reflects A4BEE's own interpretation and opinion, is not affiliated with, endorsed by, or verified with Sibylla Biotech S.p.A., and may be incomplete or inaccurate. All company names and trademarks are the property of their respective owners. To request a correction or removal, contact [email protected].