Moleculomics offer services and bespoke technology platforms to assist the processes of lead discovery, toxicology and re-purposing. Our approach is flexible and collaborative. Our company values recognise the importance of confidentiality – our services are hosted by proprietary software and project-dedicated hardware. It is our strategy to help develop and enhance intellectual property through trusted partnerships. Click the tabs below to learn more about each of the services we currently offer
3D structural modelling and annotation of proteins from sequence at individual, family and whole proteome scale The Moleculomics in silico modelling pipeline was developed over a period of 12 years. Described as “world-leading”, it predicts the 3D structure of proteins (drug receptors, transport proteins, ion channels, enzymes, antibodies, structural proteins etc.) directly from sequence.
- Accuracy – The pipeline combining the two main approaches used in protein modelling; homology modelling and protein threading which are able to reliably predict structure, including regions of low and remote homology. This facilitates accurate prediction of structural conformation, critical for understanding protein function and potential molecular interactions. The resulting quality of these models are critical to all subsequent workflows.
- Mutations and nsSNPs – A significant capacity is underpinned by our track record in structural and functional analysis of non-synonymous single nucleotide polymorphisms (nsSNPs). The resulting data may be used to analyse how an individual responds to a given drug, which is essential knowledge for future delivery of healthcare and the development of personalised medicine. We have structurally characterised the impact of nsSNPs in many microbial and eukaryotic organisms, including human, which have featured in over 20 publications since 2009 in collaboration with laboratory researchers across the world (several of which were reviewed in Mullins JGL. (2012) Structural modelling pipelines in next generation sequencing projects. Adv Protein Chem Struct Biol. 2012;89:117-67.)
- Quality assessment – The system has an integrated quality assessment check and protein clipping tool. This was developed in response to regions of remote homology affecting results. Long chains of amino acids with no definable structure affect results producing exceptionally high binding affinities, which are not representative of the real life situation and as a result generate infeasible false positives. Moleculomics apply an in-house developed protein “clipper” program to remove unstructured regions of no homology to any determined structure. The process is augmented with a quality check of the final conformational structure and residue by residue geometry.
Moleculomics offer automated protein modelling services which are a fast and efficient method of obtaining high quality structural models. In addition, Moleculomics have several technical capabilities which are offered on a bespoke basis:
- Protein annotation and functional prediction – One of the most substantial generic challenges facing researchers using the powerful new technologies of whole genome sequencing is reliably assessing the functional impact of proteins. We offer the technical capability to predict the function of proteins through Structural Activity Relationships and the functional impact of genetic differences upon protein function, and therefore the expressed activity of our cells, by high performance computing of changes in protein structure associated with known and newly discovered variants identified by whole genome sequencing projects.
- Homology testing (to identify areas of similar or different structure) – We have developed and tested in conjunction with the Centre for Defence Enterprise a novel sifting algorithm that clusters proteins by reference to the homology modelling templates used in the model and are able to applying this filtering by template approach to vast datasets. This is a powerful tool for drug development and repositioning but also a method of identifying structural homology across multiple species, to identify the relevance of proposed clinical trials.
- Assessment of resistance and protein evolution – Our capability in modelling resistance and evolution is closely linked to that for modelling mutations. A novel approach of structural alignment enables the test sequence to be aligned with homologous structural templates and the highest homology fit for a particular section of sequence is identified, and the corresponding secondary structure from the template is adopted. Along the length of a given sequence, a number of templates are typically used, normally 3-12 in the construction of an entire protein. Even minor changes in sequence can result in the adoption of a different template for a given section. This “template interchange” is the driver for fold changes observed in the final structures of variant proteins. Recording the changes in template coverage between wild type and variant proteins provides a rapid indication and measure of structural change. This capability has been developed since 2011, originally applied to the modelling of the emergence of azole resistance in fungi mediated by a similar P450 protein to CYP2D6 (Mullins et al., 2011 PLoS One. 2011;6(6):e20973), and also in our stratification of genotype-phenotype correlations in hyperekplexia, apnoeas, learning difficulties and speech delay in humans (Thomas et al., 2013 Brain 136(Pt 10):3085-95).
High throughput affinity screening and docking of compounds against panels of proteins of known toxicity to identify and eliminate problem compounds earlier within the R&D cycle
The prevailing approach to active molecule discovery in recent decades is that of rational molecular design, involving the development of new ligands for a specific target (normally a specific site in a single protein or a small number of targets that bind with the ligand). Adverse events due to off-target interactions are not widely assessed by in silico approaches. In vitro approaches are increasingly employed, typically involving toxicology panels comprising binding assays for around 44 protein receptors [Bowes et al., (2012)]. Such in vitro toxicology panel data provide valuable validation information for in silico approaches.
Moleculomics offer an in silico screening platform that will enable analysis of the predicted interaction between any compound and panels of selected target proteins. Two such panels are offered which may be used at any point during the R&D cycle, although best suited to preclinical or phase zero pharmaceutical profiling:
Panel 44 – A target panel of 44 receptors adopted as the industry standard first step in the drug discovery process, as recommended by four major pharmaceutical companies (Joanne Bowes and colleagues).
Panel 331 – A panel of similar concept to Panel 44, this time based upon the in silico screening of 331 enzymatic and receptor signalling assays featured in Sipes et al., 2013 and collated in the ToxCast database.
The products described above offer the following benefits:
- A wealth of extremely valuable molecular knowledge which enables a paradigm shift from reliance on observing effects at cellular (in vitro) and animal (in vivo) level, to predicting effects based on an understanding of the whole system at the molecular level.
- A reduction in both drug development time and costs through the ability to screen for toxic or adverse reactions to enable users to identify and eliminate problem compounds sooner within the R&D cycle.
- Substantially reduced lead times and lower screening costs to the equivalent in vitro
- A reduced reliance upon animal testing.
This quantitative simulation method offers significant savings in time, money and material over expensive in vitro and in vivo studies as the first step in the drug discovery process for the identification of the most problematic off-target interactions.
 Bowes et al (2012), Nature reviews: Drug Discovery, 11: 909-922
 Profiling 976 ToxCast Chemicals across 331Enzymatic and Receptor Signaling Assays – Sipes et al 2013.
Panel 44 Data Sheet, Panel 331 Data Sheet
High throughput affinity screening and docking of specified compounds against any of 47 tissue specific panels of proteins to provide molecular and structural information to evaluate compound efficacy and toxicity
Drugs fail for a number of reasons. Challenged fundamentally by a lack of efficacy and/or high toxicity, the underlying problem is that interactions between proteins and chemical compounds are highly complex.
Moleculomics have developed tools to assist the development of compounds by supplementing the R&D process with information relating to the structural and functional interactions of compounds with whole tissue panels.
This service provides predicted affinities for any given compound or libraries of compounds through assessment of the affinity of a given interaction, normalised to an extensive validation exercise involving the high throughput screening of vast libraries to calculate binding affinity “hit thresholds”. This system offers a proven and extensively validated process which not only predicts hits, but also supplements R&D processes with data such as the 3D structural models, and assessment of the binding site to confirm whether or not a ligand is bound at the normal activation site or an alternative docking site. This approach is able to rationalise both silent antagonists and agonists whereby ligands are predicted to bind with the target, but not activating the associated pathway. Finally, Structure-Activity relationship data is provided for each hit, where possible.
Moleculomics offer 47 separate tissue panels which are equally powerful in the understanding of both efficacious and toxic effects. Popular panels include; skin and eye for the assessment of toxic effects and also the liver panel to assess and understand the metabolism of the compound.
Screening novel compounds against specific tissue panels provides an efficient and cost effective method to accelerate the R&D process, however this fails to predict the interaction of the compound/s across the whole proteome which is critical in understanding off target interactions.
High throughput affinity screening and docking of compounds against whole proteomes to provide comprehensive molecular and structural information to evaluate compound efficacy and toxicity
The ability to predict the interaction between a limited number of proteins and compounds provides a wealth of structural and functional information to assist drug design, however it fails to realise the full potential of in silico approaches. Moleculomics have developed innovative high throughput algorithms to provide a comprehensive screen against the whole human proteome.
This approach provides information pertaining to every “hit” within the human proteome, which when analysed in conjunction with Pathway Analysis provides full knowledge of potential adverse outcomes at whole proteome scale.
Annotated Protein-Protein Interaction data pertaining to link Molecular Initiation Events with understanding of Modes of Action and tissue specific pathway analysis
Moleculomics offer Pathway Analysis tools to facilitate the understanding of molecular events from the Molecular Initiation Event (the first interaction of a chemical with a biological system) through a series of events to identify the mechanism of a given outcome; efficacious or toxic.
This approach advances compound development processes from assessment of traditional endpoints to mechanistic understanding of chemical interactions at a molecular level.
At whole proteome level, pathway analysis spans multiple levels of biological organisation from the initial exposure to a chemical through to the systemic outcome(s).
This tool is offered as an extension to tissue specific screening and/or whole proteome screening (link to each) to facilitate comprehensive understanding, not only of the interaction of a chemical with a receptor, but an understanding of the associated consequences; the activation or deactivation of specific pathways.
These novel approaches will need to advance the design of currently available predictive tools to enable:
- Screening of chemicals for interaction with numerous receptors.
- Identification, qualitative and quantitative assessment of receptor binding.
- Understanding of the (potential) consequences of binding (e.g. is binding likely to result in an adverse event?).
- Improved, human-relevant safety assessment across the bioscience sector.
The current Pathway Analysis tool is focussed upon toxicology studies. A similar tool is currently being developed to identify efficacious pathways, linked to disease networks which will be released soon…
Prepopulated platforms of protein-ligand or protein-protein interactions developed to your specification providing high value molecular and structural information, integrated to your existing workflow
Moleculomics appreciate the limitations of outsourcing and understand that a “black box” solution is not always required. For this reason we have created an interface to allow easy integration into your workflow to empower your researchers to extract answers to the questions of relevance to your R&D programmes.
The interface is designed to accommodate any of the Moleculomics services described in Service Offering. The system is fully customisable, the full potential of the system is indicated below:
Human3DProteome – whole system drug discovery, drug repurposing and toxicity screening
Human3DProteome is a breakthrough technology for rapid discovery of better and safer drug compounds – a comprehensive in silico platform of the entire human proteome (all the receptors, enzymes etc., encoded by the human genome), providing a unique molecular level structural database including detailed characterisation of active sites and millions of protein-compound interactions.
It is a powerful open-ended, virtual lead discovery environment, and the only in silico platform that allows for screening against all receptor drug targets and for off-target interactions of a compound against the whole proteome.
Human3DProteome is available as a licensed web or installed platform service and offers the following product options, either individually or together:
Human3D Drug Discovery Pipeline – the molecular docking pipeline is applied to user-specified query compounds from small scale to extremely high throughput and employs machine learning approaches for widening discovery of protein-compound interactions involving comprehensive similarity searching of biological and chemical space, considering high chemical diversity across the entire proteome. Drug-target hits are linked by our in vitro knowledgebase to downstream metabolic and signalling pathways and to specific therapeutic areas.
Human3D Drug Discovery Database –The continually updated structure-function database provides thousands of pre-run “clean lead” compounds, pharmacophores and their conformers screened for interaction with more than 1,600 known drug targets, including 750 with known pharmacological action. Drug-target hits are linked by our in vitro knowledgebase to downstream metabolic and signalling pathways and to specific therapeutic areas. Compounds of interest may be run through the Human3D Drug Discovery Pipeline.
Human3D Drug Repurposing Platform – specialised capability for drug repurposing with validated multi-method dockings of all 1,600 FDA approved drugs to known targets, and tens of thousands of new repurposing leads, continually updated.
Human3D Toxicity Screen/span> – providing early indication of potential toxicity of compounds by in silico molecular docking with a selection of specialist panels or every protein in the human body. This approach is fully complementary with SAR approaches but provides greater coverage of biological space.