The occurrence of modular peptide repeats in load-bearing (structural) proteins is common in nature, with distinctive peptide sequences that often remain conserved across different phylogenetic lineages. These highly conserved peptide sequences endow specific mechanical properties to the material, such as toughness or elasticity. Here, using bioinformatic tools and phylogenetic analysis, we have identified the GX8 peptide with the sequence GLYGGYGX (where X can be any residue) in a wide range of organisms. By simple mutation of the X residue, we demonstrate that GX8 can be self-assembled into various supramolecular structures, exhibiting vastly different physicochemical and viscoelastic properties, from liquid-like coacervate microdroplets to hydrogels to stiff solid materials. A combination of spectroscopic, electron microscopy, mechanical, and molecular dynamics studies is employed to obtain insights into molecular scale interactions driving self-assembly of GX8 peptides, underscoring that π−π stacking and hydrophobic interactions are the drivers of peptide self-assembly, whereas the X residue determines the extent of hydrogen bonding that regulates the macroscopic mechanical response. This study highlights the ability of single amino-acid polymorphism to tune the supramolecular assembly and bulk material properties of GX8 peptides, enabling us to cover a broad range of potential biomedical applications such as hydrogels for tissue engineering or coacervates for drug delivery.

No abstract

While there has been significant progress in molecular property prediction in computer-aided drug design, there is a critical need to have fast and accurate models. Many of the currently available methods are mostly specialists in predicting specific properties, leading to the use of many models side-by-side that lead to impossibly high computational overheads for the common researcher. Henceforth, the authors propose a single, generalist unified model exploiting graph convolutional variational encoders that can simultaneously predict multiple properties such as absorption, distribution, metabolism, excretion and toxicity (ADMET), target-specific docking score prediction and drug-drug interactions. Considerably, the use of this method allows for state-of-the-art virtual screening with an acceleration advantage of up to two orders of magnitude. The minimisation of a graph variational encoder's latent space also allows for accelerated development of specific drugs for targets with Pareto optimality principles considered, and has the added advantage of explainability.