Acellera Ltd

Fast prediction of the interaction between a set of ligands and major human biological and signaling pathways (KEGG and Reactome databases) using state-of-the-art self-normalizing neural networks. Main use for repurposing, deorphaning, off target prediction.

All simulations are run on Acellera infrastructure. CDA can be signed upon customer needs. All IP generated during a service (results of the simulations) is property of the customer.

Determine binding hotspots (catalytic, allosteric and hidden sites)  on your target using mixed-solvent molecular dynamics simulations in presence of small probes such as benzene, isopropanol or imidazol. Cryptoscout runs simulations on your protein in a mixture with small probes that scan the protein surface to predict cryptic sites and druggable cavities.

Predict the absolute binding affinity (dG) of a set of ligands experimentally resolved or docked in a protein using a state-of-the-art neural network-based predictor, KDeep. Only known binders ligands can be evaluated by KDeep (for instance ligands after functional assays). KDeep has been succesfully used for the D3R Grand challenge 4.

All simulations are run on Acellera infrastructure. CDA can be signed upon customer needs. All IP generated during a service (results of the simulations) is property of the customer.

Virtual Screening - Discriminate between binders and non binders. BindScope performs the re-ranking of a structural virtual screening of a library of compounds against your protein of interest using a neural-network-based predictor of binding. While ideally the HTS should be provided by the costumer, we can optionally run the virtual screening using our in-house docking tools. The algorithm will classify the ligands and rank them following their calculated probability (from 0 to 1) to bind the target. Useful after a virtual screening performed by docking, it allows to rescore and improve ligands ranking.

All simulations are run on Acellera infrastructure. CDA can be signed upon customer needs. All IP generated during a service (results of the simulations) is property of the customer.

De novo design of leads molecules. Generate novel compounds from a seed compound through generative shape-based neural network decoding.

Fast service, generates hundreds of molecules per second. Requires filter definition to select compounds of interest.

All simulations are run on Acellera infrastructure. CDA can be signed upon customer needs. All IP generated during a service (results of the simulations) is property of the customer.

Understand protein-ligand interactions with all atom resolution using our high-throughput molecular dynamics technology. Obtain binding poses, kinetics, binding affinities and binding pathways for your compounds of choice.

Probe the binding properties of your fragment hits. Understand their binding pose, and affinity, as well as the specificity and selectivity of the interactions between the ligands and the target receptor.

The deliverables are:

• Target conformational space study
• Binding site discovery
• Binding pose determination
• Thermodynamics of binding, dG
• Kinetics of binding, kon/koff
• Video reports

This platform for small molecule discovery is unique in that it can assign binding site (including selectivity of binding and specificity of the intereaction), and pose (of binding and any metastable states) thanks to the simulation speeds we can obtain with Metrocubo, and the skills we are developing in data analysis. No other platform can give such complete data sets, which we can also help validate experimentally via NMR. This information can be invaluable for generating high affinity compounds for any target. In fact, there are many success stories in pharma relating the power of fragment based approaches for drug discovery.

On-demand molecular dynamics simulations service on our AceCloud infrastructure. Obtain up to a millisecond of molecular dynamics (MD) data using our patent-pending AceCloud compute infrastructure. Data analysis assistance is optional.
We prepare the inputs files, run the simulations and deliver the trajectories

Conformational Analysis & pocket detection

Determine the conformer space of biomolecule including small peptide, nucleic acid, carbohydrate or larger target protein in terms of their kinetic and thermodynamic fingerprints. Detect new pocket (allosteric, cryptic). Determine structures compatible with NMR data.

Metrocubo is a highly compact hardware-software solution for running GPU accelerated operations. That includes both molecular dynamics simulations and machine learning problems. Metrocubo GPU hardware delivers the highest teraflops per dollar ratio and offers your organization an efficient entry into the era of desktop high-throughput molecular dynamics simulations and artificial intelligence. Can be customised on demand

HTMD is a customizable molecular-specific programmable environment to prepare, handle, simulate, visualize and analyse molecular systems. HTMD is based on Python, so scientists can easily extend it to their needs. With HTMD is possible to do very complex protocols in just few lines.

In a single script, it is possible to plan an entire computational experiment, from manipulating PDBs, building, executing and analysing simulations, computing Markov state models, kinetic rates, affinities and pathways.

HTMD has been successfully used for FBDD, for GPCR ligand binding, Antibody structural analysis and more.