Table 1 The major phases of the drug discovery and development value chain, with approximate timescales in the absence of AI or data science tools
From: Addressing infectious diseases in Africa by accelerating drug discovery through data science
Target identification and validation (1–2 years) | Hit identification and lead optimisation (2–5 years) | Preclinical testing (1–2 years) | Clinical trials (6–10 years) | Regulatory approval and post-marketing surveillance (1–3 years) |
|---|---|---|---|---|
• Examine genomic datasets to uncover factors contributing to diseases prevalent in Africa, to identify novel therapeutic targets and personalised treatments tailored to the genetic diversity of African populations. • Integrate analyses of biological interaction networks and chemical information to identify key genes or proteins involved in disease pathways, supporting both target discovery and validation. • Apply predictive models to estimate the ‘druggability’ of a target (i.e. the likelihood of it being successfully modulated by a small molecule) to streamline drug discovery efforts. | • Virtual screening of large compound libraries to identify molecules with promising biological activity, particularly in the case of natural products for which scaffold-hopping strategies can support synthetic accessibility. • Explore opportunities for repurposing and repositioning existing drugs for diseases that disproportionately affect African populations. • Devise synthetic routes tailored to available reagents in order to reduce costs and avoid delays linked to reagent procurement. • Employ predictive models for absorption, distribution, metabolism, excretion and toxicity (AMDET) to prioritise molecules with favourable profiles. • Generate new chemical structures that are computationally optimised for desired drug-like properties to accelerate the transition from a hit to a lead compound. • Predict the binding affinity between potential drugs and their target proteins to support optimisation efforts. | • Anticipate potential toxicities early in the development process to reduce the need for costly animal studies. • Identify biomarkers linked to treatment outcomes or disease progression, aiding preclinical testing and patient stratification. | • Predict patient responses to treatments in order to enable adaptive trial designs and optimal dosing regimens. • Assess patient datasets to match individuals to clinical trials according to genetic profiles, disease characteristics and other factors, leading to more efficient recruitment. | • Track real-world health data to detect adverse reactions or drug interactions, strengthening post-approval monitoring and ensuring patient safety. |
