Research Areas


Computer-assisted molecular design:

CAMD has emerged from recent advances in computational chemistry and computer technology, and promises to revolutionize the design of functional molecules. In the pharmaceutical industry, the holy grail of CAMD is to identify highly potent and specific drugs using only computational methods and structural information on the target protein. Our research activities in this area focus on the development of new computational tools for CAMD and their application to the discovery of new drugs and other nano-scale molecular devices.

[reverse-docking]1- Reverse-docking. We have developed “reverse-docking” as a new paradigm for the computer-assisted design of artificial receptors, chemosensors, catalysts and other nano-molecular devices. The basic approach hinges on docking large, flexible molecular receptors/catalysts around rigid ligands/transition state models. We are currently focusing on the study of asymmetric organocatalysts, metal-free catalysts capable of directing asymmetric reactions. The resulting 3D models of organocatalyzed reactions provide insight into the mode of asymmetric induction, and predict the product enantiomers in 100% of cases studied to date. Efforts are currently underway to broaden the scope of reverse-docking applications, including the virtual screening of asymmetric catalysts.

Watch video lecture: "The computational study of organocatalysis: thinking (and docking) outside the box"


2- New QM/MM tools. The ever-increasing speed of computers, coupled to the advent of large computing clusters, promises to bring high-throughput quantum mechanics calculations to the chemist's desktop. We have developed GI-MOE, a new computational tool for carrying out large-scale calculations, normally reserved to molecular mechanics, using full quantum mechanical methods.GI-MOE facilitates automated calculations on large databases of molecules, including geometry optimizations, transition-state calculations, potential scans, as well as ONIOM calculations without recourse to cumbersome command-line interaction with the programs.

3- SPLASH. Molecular docking is a computer-based method for predicting the 3D structure of a drug-receptor complex; it can potentially obviate the need to synthesize a potential drug to assess its biological activity. Our work addresses the critical yet unresolved issue of “bridging water molecules” that may occur at the interface of a drug-protein complex. We have developed SPLASH – selected poses of ligands with active site hydration – a method for docking ligands into proteins with no prior knowledge of active-site water positions. From a crystallographic protein structure, SPLASH generates several plausible active site hydration patterns then docks a ligand to each of these. Using acetylcholinesterase (AChE) inhibitors for methodology development and testing, SPLASH was able to reproduce the crystallographic binding pose for 15/15 inhibitor-AChE complexes.

4- Free energy calculations. Using the free energy perturbation (FEP) technique, the thermodynamic underpinnings of molecular recognition are probed with a series of abiotic receptors developed in our group. Relative and absolute free energies of binding are being computed, and used to probe the effect of solvent, in particular non-aqueous environments. To date, computational studies have been conducted on tripodal receptors of organic triacids and simpler U-turn receptors of nucleotide bases, two host-guest systems functioning in chloroform and binary solvent mixtures.

[docking]5- Homology modeling of SARS-CoV Mpro protein





Asymmetric organocatalysis:

organocatOur reverse-docking methodology has been used principally for the study of asymmetric organocatalysis. The method yields useful 3D models of the transition states of asymmetric organocatalyzed reactions, and successfully correlates the enantioselectivity in 100% of cases studied to date.
An interesting spinoff of the computational work has been the discovery of squaramide derivatives as organocatalysts for carrying out proton transfers, tautomerizations, and asymmetric aldol reactions under Baylis-Hillman conditions.


Bent bonds / antiperiplanar hypothesis:


In collaboration with Pierre Deslongchamps, we have put forth the bent bond/antiperiplanar (BBA) hypothesis, a qualitative yet novel theoretical model for understanding an ever-widening range of organic reactions involving double bonds and/or carbonyl groups. It takes into consideration bent bonds (tau-bonds), the antiperiplanar hypothesis, the classic theory of resonance, and the Walden inversion. The model accounts for the conformation and reactivity of alkenes, carbonyl and carboxyl derivatives, conjugated systems as well as other functional groups. The BBA hypothesis also provides a simple model to understand aromaticity, electrocyclic reactions, cycloaddition reactions, sigmatropic rearrangements, and other reactions.



Taxane chemistry:

taxolPaclitaxel (Taxol®) and docetaxel (Taxotere®), two taxane-based compounds, have been praised as the most important anticancer drugs to emerge from the pharmaceutical industry in the last 30 years. They are now widely used to treat breast, ovarian, non-small cell lung cancers, as well as Kaposi’s sarcoma, and constitute a multi-billion dollar industry. A key strategy for producing paclitaxel is that of semi-synthesis where one of its common metabolites is transformed by a chemical route. We are developing novel approaches to the semi-synthesis of pharmaceutically important taxanes.

Molecular recognition:

Recognition at the molecular level plays a central role in all biological processes. Not only does it dictate the assembly of biological structures but it also governs cellular events such as transport, regulation, communication, and enzymatic catalysis. A better understanding of the binding forces involved in molecular recognition is essential to the development of new drugs and for the design of original "molecular devices" such as chemosensors, probes, and catalysts.

triacid1- Design, synthesis and study of "abiotic receptors" (non-biological) capable of selective recognition of bio-relevant targets, and development these receptors into original molecular devices such as chemosensors and probes. First, receptors are designed by computer-assisted molecular modeling. They are then assembled via modern organic synthesis techniques. In a third phase, the binding properties of these receptors towards a variety of biologically relevant guests are evaluated, usually by high-field NMR, calorimetry, and X-ray crystallography.


yu2- Rational design of self-assembling supramolecular structures. Such self-assemblies form spontaneously either in solution or in the solid state, may display novel physico-chemical properties, and could lead to the development of new materials. The example on the left illustrates a novel bicyclic dilactam synthesized in our group which self-assembles in the solid-state in a controlled fashion.




This free Dreamweaver template created by