Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
Several challenges hamper the translation of preclinical drug discovery efforts into safe and effective therapies for psychiatric disorders. In their Perspective, Khodosevich and colleagues highlight differences between drug targets in animal models and those in patients as a key reason for failed clinical studies. They present a framework that integrates single-cell and spatial omics data to overcome this loss in translation, which they hope will assist the development of new drugs for diseases such as schizophrenia and major depressive disorder.
Advances with deep learning, the growth of databases of molecules for virtual screening and improvements in computational power have supported the emergence of a new field of quantitative structure–activity relationship (QSAR) modelling applications that Tropsha et al. term ‘deep QSAR’. This article discusses key advances in the field, including deep generative and reinforcement learning approaches in molecular design, deep learning models for synthetic planning, and the use of deep QSAR models in structure-based virtual screening.
Current treatments for the two most common chronic inflammatory skin disorders, atopic dermatitis and psoriasis, are mainly limited to the alleviation of symptoms. Here, Bieber discusses opportunities and strategies for the discovery and development of interventions aimed at inducing remission of these disorders and their associated comorbidities, assessing key issues that must be addressed to reach this goal.
The COVID-19 pandemic spurred a wave of rapid and collaborative drug discovery efforts. This Perspective article summarizes scientific drivers and considerations behind such antiviral small-molecule discovery programmes and proposes strategies to accelerate future efforts.
Bivalent protein degraders such as proteolysis targeting chimeras (PROTACs) are entering clinical trials, with a current focus on oral administration. O’Brien Laramy et al. propose that implementing non-oral drug delivery technologies guided by pharmacokinetic–pharmacodynamic modelling could expand the chemical design space for degraders as well as the number of druggable targets.
Bayesian statistical approaches that explicitly incorporate existing data into clinical trial design, analysis and decision-making have the potential to substantially reduce the time and cost of bringing new medicines to patients in some contexts, but remain underused. This Perspective highlights the value of Bayesian methods in drug development, discusses barriers to their application and recommends approaches to address them.
Investigative toxicology tools and strategies are used in pharmaceutical companies to reduce safety-related attrition in drug development. This Perspective article summarizes the key goals of investigative toxicology, highlights current approaches and discusses selected emerging technologies that have the potential to improve the current safety-testing paradigm.
