Computational life sciences, on demand.
Computational chemistry, bioinformatics, and machine learning across the drug discovery and natural products pipeline — rigorous, reproducible analysis without the overhead of a wet laboratory.
Eight pipelines across the discovery stack.
01VS · 01Virtual Screening
We evaluate large compound libraries against a biological target using structure- and ligand-based methods, filtering commercial, natural-product, or internal collections down to a prioritised hit list before any lab work begins.
Ranked hit list with binding scores, compound visualisations, filtering rationale, and summary report.
02DOCK · 02Molecular Docking
We model how a compound binds a protein target — predicting binding pose, affinity scores, and the key interactions (hydrogen bonds, hydrophobic contacts, pi-stacking) that drive potency.
Binding-mode visualisations, docking scores, interaction analysis, and a written interpretation of findings.
03MD · 03Molecular Dynamics Simulations
Where docking is a snapshot, MD captures how a protein-ligand complex behaves over time — assessing binding stability, conformational flexibility, and thermodynamics over nanosecond to microsecond timescales.
Trajectory analysis, RMSD and RMSF plots, binding free-energy estimates, and a structured results report.
04PHYTO · 04Bioactivity Prediction of Plant Compounds
Southern Africa holds one of the world's richest reservoirs of medicinal plant diversity. We predict the bioactivity of phytochemical constituents against broad target panels, grounding traditional knowledge in quantitative evidence.
Target prediction profiles, bioactivity probability scores, docking-based validation for priority compounds, and a publication-ready research summary.
05ADMET · 05In Silico ADMET Profiling
Strong target activity is worthless if a compound is poorly absorbed, toxic, or rapidly metabolised. We computationally predict Absorption, Distribution, Metabolism, Excretion, and Toxicity as an early safety and pharmacokinetic filter.
Full ADMET profile per compound, flagged liabilities, comparison across compound series, and a regulatory-aligned summary report.
06TOX · 06Predictive Toxicology Modelling
We build and apply machine-learning models on validated datasets (Tox21, ToxCast) to predict hepatotoxicity, cardiotoxicity (hERG), mutagenicity (Ames), and reproductive toxicity — fine-tuned to specific compound classes.
Per-compound toxicity predictions across specified endpoints, confidence scores, mechanistic flags, and a regulatory-formatted summary aligned to MCAZ and SAHPRA expectations.
07ECO · 07Computational Environmental Toxicology
In silico assessment of a substance's environmental risk profile — using QSAR-based approaches to predict aquatic toxicity, bioaccumulation potential, and environmental persistence across trophic levels.
Predicted environmental endpoints, species sensitivity profiles, risk-characterisation summary, and documentation formatted for regulatory review.
08NLP · 08Literature Mining & Knowledge Graphs
We deploy NLP pipelines on domain-specific biomedical models to extract structured knowledge at scale — mapping drug-target relationships, building bioactivity databases, and constructing queryable knowledge graphs.
Structured dataset or knowledge graph of extracted relationships, entity visualisations, gap-analysis report, and documentation of extraction methodology.
Evidence before the bench — not after it.
No wet-lab overhead
Rigorous, reproducible analysis delivered entirely digitally — focus experimental budget only on the candidates that survive computational triage.
Southern Africa-rooted
Built to characterise the region's medicinal plant diversity and aligned to regulatory expectations for MCAZ and SAHPRA dossier submissions.
Project or partnership
Engagements are scoped as standalone projects or ongoing research partnerships — with custom pipelines and white-label outputs available on request.
Bring us the target. We'll run the science.
Tell us the protein, the library, or the compound series — and we'll come back with a scoped pipeline, timeline, and the deliverables you'll receive. Standalone projects or ongoing research partnerships.