How exascale computing can shape drug design: A perspective from multiscale QM/MM molecular dynamics simulations and machine learning-aided enhanced sampling algorithms

Molecular simulations are an essential asset in the first steps of drug design campaigns. However, the requirement of high-throughput limits applications mainly to qualitative approaches with low computational cost, but also low accuracy. Unlocking the potential of more rigorous quantum mechanical/molecular mechanics (QM/MM) models combined with molecular dynamics-based free energy techniques could have a tremendous impact. Indeed, these two relatively old techniques are emerging as promising methods in the field. This has been favored by the exponential growth of computer power and the proliferation of powerful data-driven methods. Here, we briefly review recent advances and applications, and give our perspective on the impact that QM/MM and free-energy methods combined with machine learning-aided algorithms can have on drug design.

Effective data-driven collective variables for free energy calculations from metadynamics of paths

A variety of enhanced sampling (ES) methods predict multidimensional free energy landscapes associated with biological and other molecular processes as a function of a few selected collective variables (CVs). The accuracy of these methods is crucially dependent on the ability of the chosen CVs to capture the relevant slow degrees of freedom of the system. For complex processes, finding such CVs is the real challenge. Machine learning (ML) CVs offer, in principle, a solution to handle this problem. However, these methods rely on the availability of high-quality datasets-ideally incorporating information about physical pathways and transition states-which are difficult to access, therefore greatly limiting their domain of application. Here, we demonstrate how these datasets can be generated by means of ES simulations in trajectory space via the metadynamics of paths algorithm. The approach is expected to provide a general and efficient way to generate efficient ML-based CVs for the fast prediction of free energy landscapes in ES simulations. We demonstrate our approach with two numerical examples, a 2D model potential and the isomerization of alanine dipeptide, using deep targeted discriminant analysis as our ML-based CV of choice.

Kinetic and structural details of urease inactivation by thiuram disulphides

This paper reports on the molecular details of the reactivity of urease, a nickel-dependent enzyme that catalyses the last step of organic nitrogen mineralization, with thiuram disulphides, a class of molecules known to inactivate the enzyme with high efficacy but for which the mechanism of action had not been yet established. IC50 values of tetramethylthiuram disulphide (TMTD or Thiram) and tetraethylthiuram disulphide (TETD or Disulfiram) in the low micromolar range were determined for plant and bacterial ureases. The X-ray crystal structure of Sporosarcina pasteurii urease inactivated by Thiram, determined at 1.68 Å resolution, revealed the presence of a covalent modification of the catalytically essential cysteine residue. This is located on the flexible flap that modulates the size of the active site channel and cavity. Formation of a Cys-S-S-C(S)-N(CH3)2 functionality responsible for enzyme inactivation was observed. Quantum-mechanical calculations carried out to rationalise the large reactivity of the active site cysteine support the view that a conserved histidine residue, adjacent to the cysteine in the active site flap, modulates the charge and electron density along the thiol SH bond by shifting electrons towards the sulphur atom and rendering the thiol proton more reactive. We speculate that this proton could be transferred to the nickel-coordinated urea amide group to yield a molecule of ammonia from the generated Curea-NH3+ functionality during catalysis.

Drug Design in the Exascale Era: A Perspective from Massively Parallel QM/MM Simulations

The initial phases of drug discovery - in silico drug design - could benefit from first principle Quantum Mechanics/Molecular Mechanics (QM/MM) molecular dynamics (MD) simulations in explicit solvent, yet many applications are currently limited by the short time scales that this approach can cover. Developing scalable first principle QM/MM MD interfaces fully exploiting current exascale machines - so far an unmet and crucial goal - will help overcome this problem, opening the way to the study of the thermodynamics and kinetics of ligand binding to protein with first principle accuracy. Here, taking two relevant case studies involving the interactions of ligands with rather large enzymes, we showcase the use of our recently developed massively scalable Multiscale Modeling in Computational Chemistry (MiMiC) QM/MM framework (currently using DFT to describe the QM region) to investigate reactions and ligand binding in enzymes of pharmacological relevance. We also demonstrate for the first time strong scaling of MiMiC-QM/MM MD simulations with parallel efficiency of ∼70% up to >80,000 cores. Thus, among many others, the MiMiC interface represents a promising candidate toward exascale applications by combining machine learning with statistical mechanics based algorithms tailored for exascale supercomputers.

MiMiCPy: An Efficient Toolkit for MiMiC-Based QM/MM Simulations

MiMiC is a highly flexible, extremely scalable multiscale modeling framework. It couples the CPMD (quantum mechanics, QM) and GROMACS (molecular mechanics, MM) codes. The code requires preparing separate input files for the two programs with a selection of the QM region. This can be a tedious procedure prone to human error, especially when dealing with large QM regions. Here, we present MiMiCPy, a user-friendly tool that automatizes the preparation of MiMiC input files. It is written in Python 3 with an object-oriented approach. The main subcommand PrepQM can be used to generate MiMiC inputs directly from the command line or through a PyMOL/VMD plugin for visually selecting the QM region. Many other subcommands are also provided for debugging and fixing MiMiC input files. MiMiCPy is designed with a modular structure that allows seamless extensions to new program formats depending on the requirements of MiMiC.

Enhanced-Sampling Simulations for the Estimation of Ligand Binding Kinetics: Current Status and Perspective

The dissociation rate (k off) associated with ligand unbinding events from proteins is a parameter of fundamental importance in drug design. Here we review recent major advancements in molecular simulation methodologies for the prediction of k off. Next, we discuss the impact of the potential energy function models on the accuracy of calculated k off values. Finally, we provide a perspective from high-performance computing and machine learning which might help improve such predictions.

A modified nudged elastic band algorithm with adaptive spring lengths

We present a modified version of the nudged elastic band (NEB) algorithm to find minimum energy paths connecting two known configurations. We show that replacing the harmonic band-energy term with a discretized version of the Onsager-Machlup action leads to a NEB algorithm with adaptive spring lengths that automatically increase the resolution of the minimum energy path around the saddle point of the potential energy surface. The method has the same computational cost per optimization step of the standard NEB algorithm and does not introduce additional parameters. We present applications to the isomerization of alanine dipeptide, the elimination of hydrogen from ethane, and the healing of a 5-77-5 defect in graphene.

Metadynamics of Paths

We present a method to sample reactive pathways via biased molecular dynamics simulations in trajectory space. We show that the use of enhanced sampling techniques enables unconstrained exploration of multiple reaction routes. Time correlation functions are conveniently computed via reweighted averages along a single trajectory and kinetic rates are accessed at no additional cost. These abilities are illustrated analyzing a model potential and the umbrella inversion of NH_{3} in water. The algorithm allows a parallel implementation and promises to be a powerful tool for the study of rare events.

Friction of physisorbed nanotubes: rolling or sliding?

The structure and motion of carbon and h-BN nanotubes (NTs) deposited on graphene is inquired theoretically by simulations based on state-of-the-art interatomic force fields. Results show that any typical cylinder-over-surface approximation is essentially inaccurate. NTs tend to flatten at the interface with the substrate and upon driving they can either roll or slide depending on their size and on their relative orientation with the substrate. In the epitaxially aligned orientation we find that rolling is always the main mechanism of motion, producing a kinetic friction linearly growing with the number of walls, in turn causing an unprecedented supra-linear scaling with the contact area. A 30 degrees misalignment raises superlubric effects, making sliding favorable against rolling. The resulting rolling-to-sliding transition in misaligned NTs is explained in terms of the faceting appearing in large multi-wall tubes, which is responsible for the increased rotational stiffness. Modifying the geometrical conditions provides an additional means of drastically tailoring the frictional properties in this unique tribological system.

The Princess and the Nanoscale Pea: Long-Range Penetration of Surface Distortions into Layered Materials Stacks

The penetration of moiré out-of-plane distortions, formed at the heterogeneous interface of graphene and hexagonal boron nitride (h-BN), into the layered h-BN stack is investigated. For aligned contacts, the estimated characteristic penetration length of ∼4.7 nm suggests that even at the far surface of a ∼25 h-BN layer thick slab stacked atop the contact, a corrugation of ∼0.1 Å, well within experimental resolution, should still be clearly evident. The penetration length is found to strongly reduce with increasing misalignment angle of the graphene/h-BN junction, where the effect of thermal fluctuations conceals the moiré-induced corrugation in the bulk. These results can be rationalized by continuum elastic theory arguments for anisotropic media. Our findings, which are expected to generally apply for layered heterojunctions, may serve as a route to control the surface corrugation, adhesive properties, and tribological characteristics of two-dimensional materials.

Negative Friction Coefficients in Superlubric Graphite-Hexagonal Boron Nitride Heterojunctions

Negative friction coefficients, where friction is reduced upon increasing normal load, are predicted for superlubric graphite-hexagonal boron nitride heterojunctions. The origin of this counterintuitive behavior lies in the load-induced suppression of the moiré superstructure out-of-plane distortions leading to a less dissipative interfacial dynamics. Thermally induced enhancement of the out-of-plane fluctuations leads to an unusual increase of friction with temperature. The highlighted frictional mechanism is of a general nature and is expected to appear in many layered material heterojunctions.

Robust microscale superlubricity in graphite/hexagonal boron nitride layered heterojunctions

Structural superlubricity is a fascinating tribological phenomenon, in which the lateral interactions between two incommensurate contacting surfaces are effectively cancelled resulting in ultralow sliding friction. Here we report the experimental realization of robust superlubricity in microscale monocrystalline heterojunctions, which constitutes an important step towards the macroscopic scale-up of superlubricity. The results for interfaces between graphite and hexagonal boron nitride clearly demonstrate that structural superlubricity persists even when the aligned contact sustains external loads under ambient conditions. The observed frictional anisotropy in the heterojunctions is found to be orders of magnitude smaller than that measured for their homogeneous counterparts. Atomistic simulations reveal that the underlying frictional mechanisms in the two cases originate from completely different dynamical regimes. Our results are expected to be of a general nature and should be applicable to other van der Waals heterostructures.

Nanoserpents: Graphene Nanoribbon Motion on Two-Dimensional Hexagonal Materials

We demonstrate snake-like motion of graphene nanoribbons atop graphene and hexagonal boron nitride ( h-BN) substrates using fully atomistic nonequilibrium molecular dynamics simulations. The sliding dynamics of the edge-pulled nanoribbons is found to be determined by the interplay between in-plane ribbon elasticity and interfacial lattice mismatch. This results in an unusual dependence of the friction-force on the ribbon's length, exhibiting an initial linear rise that levels-off above a junction-dependent threshold value dictated by the pre-slip stress distribution within the slider. As part of this letter, we present the LAMMPS implementation of the registry-dependent interlayer potentials for graphene, h-BN, and their heterojunctions that were used herein, which provides enhanced performance and accuracy.

Blood Levels of IGF-I in Non-Small Cell Lung Cancer: Relation to Clinical Data

Recent observations have demonstrated that somatomedins, mainly insulin-like growth factor-I (IGF-I), are growth factors for non-small cell lung cancer (NSCLC). On the basis of this evidence, a study was started to evaluate serum levels of IGF-I in a group of untreated NSCLC patients. The study included 46 patients, 25 of whom had an operable tumor, while the other 21 showed distant organ metastases. IGF-I and GH serum levels were measured by RIA in each patient; moreover, in operable patients, hormonal detections were made either before, or 7 days after surgery. The control group comprised 38 age-matched healthy subjects. Mean serum levels of IGF-I were significantly higher in cancer patients with respect to controls, while no difference was seen in mean GH values. Moreover, patients with metastases showed significantly higher levels of IGF-I than the patients without. Within the operable group, patients with lung adenocarcinoma had higher levels of IGF-I than those with epidermoid cell carcinoma, but this difference was not significant. Finally, no significant difference in IGF-I mean values was seen before and after surgical removal of tumors. This preliminary study shows that NSCLC patients may present abnormally high levels of IGF-I. Because of the stimulating role of IGF-I on NSCLC growth, this evidence could play a role in the clinical course of neoplastic lung disease.

Sliding friction of graphene/hexagonal -boron nitride heterojunctions: a route to robust superlubricity

The origin of ultra-low friction exhibited by heterogeneous junctions of graphene and hexagonal boron nitride (h-BN) is revealed. For aligned interfaces, we identify a characteristic contact size, below which the junction behaves like its homogeneous counterparts with friction forces that grow linearly with the contact area. Superlubricity sets in due to the progressive appearance of Moiré patterns resulting in a collective stick-slip motion of the elevated super-structure ridges that turns into smooth soliton-like gliding with increasing contact size. Incommensurability effects are enhanced in misaligned contacts, where the friction coefficients further drop by orders of magnitude. Our fully atomistic simulations show that the superlubric regime in graphene/h-BN heterostructures persists up to significantly higher loads compared to the well-studied twisted homogeneous graphene interface. This indicates the potential of achieving robust superlubricity in practical applications using two-dimensional layered materials heterojunctions.

Friction boosted by equilibrium misalignment of incommensurate two-dimensional colloid monolayers

Colloidal two-dimensional monolayers sliding in an optical lattice are of recent importance as a frictional system. In the general case when the monolayer and optical lattices are incommensurate, we predict two important novelties, one in the static equilibrium structure, the other in the frictional behavior under sliding. Structurally, realistic simulations show that the colloid layer should possess in full equilibrium a small misalignment rotation angle relative to the optical lattice, an effect so far unnoticed but visible in some published experimental moiré patterns. Under sliding, this misalignment has the effect of boosting the colloid monolayer friction by a considerable factor over the hypothetical aligned case discussed so far. A frictional increase of similar origin must generally affect other incommensurate adsorbed monolayers and contacts, to be sought out case by case.

Static friction scaling of physisorbed islands: the key is in the edge

The static friction preventing the free sliding of nanosized rare gas solid islands physisorbed on incommensurate crystalline surfaces is not completely understood. Simulations modeled on Kr/Pb(111) highlight the importance and the scaling behavior of the island's edge contribution to static friction.

Posterior spinal instrumentation: biomechanical study on the role of rods on hardware response to axial load

Posterior spinal instrumentation is frequently used for the treatment of spine disorders. Importantly, different requirements have to be considered for the optimal use of these systems in various clinical scenarios. In this work, we focused on the role of rods diameter on hardware's stiffness. For this purpose, we established an in vitro model and compared the response to axial load of a posterior stabilization system, characterized by rods of different diameter (4, 5, 6 mm), with that of Dynesys®. Intuitively, the higher the stiffness of the hardware, the lower the load is transferred to the disc. However, the 4 hardware tested showed a different trend in the response to the load regimens: when increasing the load, more flexible systems display a progressive reduction in the percentage of load which is transferred to the disc while more rigid system display the opposite trend. Considering that the load which is transferred, and not by-passed by the hardware, influences the healing of a fracture; the integration of a bone graft or a cage; the fusion process, these data have a relevant impact on clinical practice and highlight features that have to be considered in the choice for the optimal posterior spinal instrumentation.

Surgical treatment of far lateral lumbar disc herniation. Identification of compressed root and discectomy by lateral approach

STUDY DESIGN

A new method is described of compressed root identification and discectomy for extraforaminal disc herniation, by a lateral intertransversalis approach.

OBJECTIVES

To describe a safe surgical approach that does not require resection of adjacent bone structures during extraforaminal discectomy.

SUMMARY OF BACKGROUND DATA

Most earlier series have reported approaches that damaged bordering bone structures with wide laminoarthrectomy. This is an attempt at a safer, simpler surgical approach.

METHODS

Thirteen patients with lateral hernia have undergone this surgical procedure since 1995. Herniectomy was performed after identification of the compressed root within the iliopsoas muscle.

RESULTS

All the patients resumed the upright position with the aid of semirigid brace 24 hours after surgery. Upon awakening from the anesthesia, no patient reported peripheral pain. Motor deficits resolved after physical rehabilitation in all but one patient. At a mean follow-up of 14 months, there was no report of back pain.

CONCLUSION

The procedure described in this article offers a simple alternative to the valid procedures presently at hand. It offers the advantage of no bone resection and of minimizing nerve structures manipulation.

Postlaminectomy cervical dislocation in von Recklinghausen's disease. A case report

STUDY DESIGN

A case of a 9-year-old girl with von Recklinghausen's disease who has acute tetraparesis caused by a complete dislocation of C6-C7 after a cervical laminectomy performed at another hospital.

OBJECTIVES

To demonstrate the treatment difficulties of cervical spine abnormality associated with neurofibromatosis.

SUMMARY OF BACKGROUND DATA

Craig and Govender have reported cases of neurofibromatosis of the cervical spine (1992).

METHODS

The patient underwent emergency surgery starting with a posterior release of the articular facets and with the positioning of two Roy-Camille plates. The dislocation of C6-C7 was reduced by an anterior approach. Finally the Roy-Camille plates were removed, a plate was implanted posteriorly at the C6 and C7 joints, and a posterior cable was positioned around the C2-C3 posterior arches.

RESULTS

More than 4.5 years after surgery, the patient's neurologic condition remains satisfactory. She can walk and run and has no sphincter disorders.

CONCLUSIONS

The results underline the importance of simultaneous anterior and posterior surgical approaches in this disease.

Relation between Surgery-Induced Prolactin Increase and the Menstrual Cycle Phase at Time of Surgery in Premenopausal Breast Cancer

It has been suggested that both the menstrual cycle phase and postoperative changes in prolactin (PRL) secretion at the time of surgery may influence the prognosis of breast cancer. The present study was carried out to evaluate the relation between menstrual cycle period and surgery-induced PRL variations. We evaluated 32 premenopausal women with operable breast carcinoma; 17 were in perimenstrual phase (days 1-6 and 21-28) and 15 were in the mid-cycle (days 7-20) period at’ the time of surgery. To investigate serum levels of PRL, venous blood samples were collected before and 7 days after surgery. Postoperative hyperprolactinemia occurred in 17/32 patients and it was statistically more frequent in patients surgically treated during the perimenstrual phase than in the mid-cycle phase (12/17 vs 5/15; p < 0.05), while no other parameter (including axillary node and estrogen receptor status) showed a significant influence on hyperprolactinemia rate. The results suggest that in premenopausal breast cancer patients surgery-induced hyperprolactinemia may be influenced by the menstrual cycle phase at the time of surgery.

Blood levels of IGF-I in non-small cell lung cancer: relation to clinical data

Recent observations have demonstrated that somatomedins, mainly insulin-like growth factor-I (IGF-I), are growth factors for non-small cell lung cancer (NSCLC). On the basis of this evidence, a study was started to evaluate serum levels of IGF-I in a group of untreated NSCLC patients. The study included 46 patients, 25 of whom had an operable tumor, while the other 21 showed distant organ metastases. IGF-I and GH serum levels were measured by RIA in each patient; moreover, in operable patients, hormonal detections were made either before, or 7 days after surgery. The control group comprised 38 age-matched healthy subjects. Mean serum levels of IGF-I were significantly higher in cancer patients with respect to controls, while no difference was seen in mean GH values. Moreover, patients with metastases showed significantly higher levels of IGF-I than the patients without. Within the operable group, patients with lung adenocarcinoma had higher levels of IGF-I than those with epidermoid cell carcinoma, but this difference was not significant. Finally, no significant difference in IGF-I mean values was seen before and after surgical removal of tumors. This preliminary study shows that NSCLC patients may present abnormally high levels of IGF-I. Because of the stimulating role of IGF-I on NSCLC growth, this evidence could play a role in the clinical course of neoplastic lung disease.

Postoperative hyperprolactinaemia and early recurrence rate in breast cancer

Serum levels of prolactin before and after surgery were measured in 90 women with breast cancer until the 5th postoperative month. Surgery-induced hyperprolactinaemia occurred in 51 patients, without significant correlation to any other clinical variable. After a median follow-up of 39 months, irrespective of each other variable (i.e. nodal involvement, oestrogen receptor status, adjuvant therapies), patients with postoperative hyperprolactinaemia had a significantly lower recurrence rate than those in whom surgery was not followed by an abnormal increase in prolactin secretion (3/51 vs. 13/39, P less than 0.001). These results suggest that, despite the stimulatory role of prolactin on mammary tumours, the lack of postoperative hyperprolactinaemia is an unfavourable prognostic factor because of its association with a higher relapse rate.