Motivation Root mean square deviation (RMSD) is one of the most useful and straightforward features for structural comparison between different conformations of the same molecule. Commonly, protein-ligand docking programs have included some utilities that allow the calculation of this value; however, they only work efficiently when exists a complete atom label equivalence between the evaluated conformations. Results We present LigRMSD, a free web-server for the automatic matching and RMSD calculations among identical or similar chemical compounds. This server allows the user to submit only a pair of identical or similar molecules or dataset of similar compounds to compare their three-dimensional conformations. Availability and implementation LigRMSD can be freely accessed at https://ligrmsd.appsbio.utalca.cl. Supplementary information Supplementary data are available at Bioinformatics online.

We present here a complete detailed theoretical analysis of l-arginine hydrolysis catalyzed by human arginase I (HARGI). Our study combines classical molecular dynamics (MD) simulations of different models for the enzyme–substrate complex and a detailed exploration of the reaction mechanism for the most plausible model using hybrid quantum mechanics/molecular mechanics techniques. Different enzyme–substrate models were built considering that the nucleophile in charge of the initial attack to the substrate could either be a water molecule or a hydroxide anion bridging between the two manganese­(II) cations present in the active site. In contrast, for the substrate, we considered four different possibilities: one with the guanidino group of the substrate protonated and three of them where this group is neutral. In this last case, we considered two different tautomeric states and two possible coordination modes with the divalent ions. Our MD simulations revealed that the most stable complexes correspond to those having a hydroxide anion as the nucleophile, while both a protonated and a neutral form of the guanidino group can bind into the active site. Our analysis of the potential energy surface reveals a complex reaction pathway, where the initial attack of the nucleophile is followed by the inversion of the epsilon nitrogen atom of the guanidino group. It produces a reaction intermediate, stabilized by means of an H−π interaction with a histidine residue, that shows high structural similitude with some of the most potent inhibitors known for HARGI (those derived from the 2­(S)-amino-6-boronohexanoic acid). Subsequently, a proton is transferred from the nucleophile to the leaving group via Asp128. The final step corresponds to the separation of the two reaction fragments: l-ornithine and urea. Our analysis is shown to be robust by comparison among the results obtained with different initial structures and functionals and also because of the good agreement with the experimental estimations of the activation energy and with mutagenesis analysis. Finally, the analysis of the electronic distribution demonstrates that manganese ions are not involved in charge-transfer processes during the reaction pathway, playing a structural and electrostatic role to stabilize the nucleophile and intermediate states of the reaction.

Cysteine proteases are an important target for the development of inhibitors that could be used as drugs to regulate the activity of these kinds of enzymes involved in many diseases, including COVID-19. For this reason, it is important to have methodological tools that allow a detailed study of their activity and inhibition, combining computational efficiency and accuracy. We here explore the performance of different quantum mechanics/molecular mechanics methods to explore the inhibition reaction mechanism of the SARS-CoV-2 3CL protease with a hydroxymethyl ketone derivative. We selected two density functional theory (DFT) functionals (B3LYP and M06-2X), two semiempirical Hamiltonians (AM1d and PM6), and two tight-binding DFT methods (DFTB3 and GFN2-xTB) to explore the free energy landscape associated with this reaction. We show that it is possible to obtain an accurate description combining molecular dynamics simulations performed using tight-binding DFT methods and single-point energy corrections at a higher QM description. The use of a computational strategy that provides reliable results at a reasonable computational cost could assist the in silico screening of possible candidates during the design of new drugs directed against cysteine proteases.

The current work provides a study on the cis–trans isomerization behaviour of N-alkylated peptides decorated with phosphonate ester groups.