PLAS-20k: Extended Dataset of Protein-Ligand Affinities from MD Simulations for Machine Learning Applications.

Computing binding affinities is of great importance in drug discovery pipeline and its prediction using advanced machine learning methods still remains a major challenge as the existing datasets and models do not consider the dynamic features of protein-ligand interactions. To this end, we have developed PLAS-20k dataset, an extension of previously developed PLAS-5k, with 97,500 independent simulations on a total of 19,500 different protein-ligand complexes. Our results show good correlation with the available experimental values, performing better than docking scores. This holds true even for a subset of ligands that follows Lipinski's rule, and for diverse clusters of complex structures, thereby highlighting the importance of PLAS-20k dataset in developing new ML models. Along with this, our dataset is also beneficial in classifying strong and weak binders compared to docking. Further, OnionNet model has been retrained on PLAS-20k dataset and is provided as a baseline for the prediction of binding affinities. We believe that large-scale MD-based datasets along with trajectories will form new synergy, paving the way for accelerating drug discovery.

PLAS-5k: Dataset of Protein-Ligand Affinities from Molecular Dynamics for Machine Learning Applications.

Computational methods and recently modern machine learning methods have played a key role in structure-based drug design. Though several benchmarking datasets are available for machine learning applications in virtual screening, accurate prediction of binding affinity for a protein-ligand complex remains a major challenge. New datasets that allow for the development of models for predicting binding affinities better than the state-of-the-art scoring functions are important. For the first time, we have developed a dataset, PLAS-5k comprised of 5000 protein-ligand complexes chosen from PDB database. The dataset consists of binding affinities along with energy components like electrostatic, van der Waals, polar and non-polar solvation energy calculated from molecular dynamics simulations using MMPBSA (Molecular Mechanics Poisson-Boltzmann Surface Area) method. The calculated binding affinities outperformed docking scores and showed a good correlation with the available experimental values. The availability of energy components may enable optimization of desired components during machine learning-based drug design. Further, OnionNet model has been retrained on PLAS-5k dataset and is provided as a baseline for the prediction of binding affinities.

Dipolar relaxation in thin films of supramolecular stacks of benzenecarboxamides and insights to enhance their ferroelectric characteristics.

The relationship between molecular structure and ferroelectric behaviour of thin films is explored in an all-organic supramolecular polymer material based on benzenecarboxamides, using atomistic molecular dynamics simulations. While increasing the number of amide groups around the phenyl core increases the dipole density of a molecule, increasing the length of the corresponding alkyl groups decreases the same. The interplay between these two contributions displays a rich behaviour on key material characteristics, in particular, the polarisation retention time. The latter is shown to be inversely proportional to the alkyl chain length, a consequence of weaker interactions between macrodipoles of stacks. Polarisation retention time was observed to be the highest in a molecule with five amide groups around the aromatic phenyl core which is explained as due to the large barrier for amide group rotation, which is one of the crucial channels for dipolar relaxation. Simulations also demonstrate that the barrier, however, does not affect the switchability of polarization, upon field reversal.

Bioinspired, ATP-driven co-operative supramolecular polymerization and its pathway dependence.

A bio-inspired, ATP-driven nucleation growth assembly is demonstrated using an amphiphilic naphthalene diimide (NDI) derivative appended with guanidinium receptors to promote specific salt-bridge type interaction with nucleotide phosphates. Detailed spectroscopic and microscopic probing revealed a pathway-dependent co-operative self-assembly to yield two-dimensional and scrolled nano-tubular bilayer assemblies under kinetic and thermodynamic conditions, respectively.

Differentiating the mechanism of self-assembly in supramolecular polymers through computation.

The mechanism by which monomers in solution, beyond a certain concentration or below a certain temperature, self-assemble to form one dimensional supramolecular polymers determines much of the bulk properties of the polymer. The two distinct pathways of assembly, namely isodesmic and cooperative, can be experimentally identified using spectroscopy and in simulations via a determination of the dependence of the association constant on the oligomer size. Employing large scale free energy calculations, we have been able to show the independence of the free energy change of oligomerization on size in the self-assembly of a [2.2]paracyclophane-tetracarboxamide ([2.2]pCpTA) derivative ([2.2]pCpTA-hex), which is experimentally shown to follow the isodesmic pathway. In contrast, simulations show the free energy change in the case of benzene-1,3,5-tricarboxamide (BTA) to depend on the oligomer size which is a signature of its cooperative nature of self-assembly. These observations are rationalized through the development of a macrodipole moment in BTA oligomers and lack thereof in the [2.2]pCpTA system.

Biomimetic temporal self-assembly via fuel-driven controlled supramolecular polymerization.

Temporal control of supramolecular assemblies to modulate the structural and transient characteristics of synthetic nanostructures is an active field of research within supramolecular chemistry. Molecular designs to attain temporal control have often taken inspiration from biological assemblies. One such assembly in Nature which has been studied extensively, for its well-defined structure and programmable self-assembly, is the ATP-driven seeded self-assembly of actin. Here we show, in a synthetic manifestation of actin self-assembly, an ATP-selective and ATP-fuelled, controlled supramolecular polymerization of a phosphate receptor functionalised monomer. It undergoes fuel-driven nucleation and seeded growth that provide length control and narrow dispersity of the resultant assemblies. Furthermore, coupling via ATP-hydrolysing enzymes yielded its transient characteristics. These results will usher investigations into synthetic analogues of important biological self-assembly motifs and will prove to be a significant advancement toward biomimetic temporally programmed materials.

Molecular modelling of supramolecular one dimensional polymers.

Supramolecular polymers exemplify the need to employ several computational techniques to study processes and phenomena occuring at varied length and time scales. Electronic processes, conformational and configurational excitations of small aggregates of chromophoric molecules, solvent effects under realistic thermodynamic conditions and mesoscale morphologies are some of the challenges which demand hierarchical modelling approaches. This review focusses on one-dimensional supramolecular polymers, the mechanism of self-assembly of monomers in polar and non-polar solvents and properties they exhibit. Directions for future work are as well outlined.

Supramolecular Polymerization of N,N',N″,N‴-tetra-(Tetradecyl)-1,3,6,8-pyrenetetracarboxamide: A Computational Study.

The role of molecular dipole orientations and intermolecular interactions in a derivative of pyrene on its supramolecular self-assembly in solution has been investigated using quantum chemical and force field based computational approaches. Five possible dipole configurations of the molecule have been examined, among which the one in which adjacent dipole vectors are antiparallel to each other is determined to be the ground state, on electrostatic grounds. Self-assembly of this molecule under realistic conditions has been studied using MD simulations. Dipolar relaxation in its liquid crystalline (LC) phase has been investigated and contrasted against that in the well-established benzene-1,3,5-tricarboxamide (BTA) family. The dihedral barrier related to the amide dipole flip is larger in the pyrene system than in BTA which explains the differences in their dipolar relaxation behaviors. The mechanism underlying polarization switching upon the application of an external electric field in the LC phase is investigated. Unlike in BTA, this switching is not associated with a reversal of the helical sense of the hydrogen bonded chains, due to differences in molecular symmetry. The observations enable general conclusions on the relationship between electric field induced chiral enhancement and symmetry to be drawn.