A Cloud Computing Platform for Scalable Relative and Absolute Binding Free Energy Predictions: New Opportunities and Challenges for Drug Discovery

Abstract

Free energy perturbation (FEP) has become widely used in drug discovery programs for binding affinity prediction between candidate compounds and their biological targets. However, limitations of FEP applications also exist, including, but not limited to, high cost, long waiting time, limited scalability, and breadth of application scenarios. To overcome these problems, we have developed XFEP, a scalable cloud computing platform for both relative and absolute free energy predictions using optimized simulation protocols. XFEP enables large-scale FEP calculations in a more efficient, scalable, and affordable way, for example, the evaluation of 5000 compounds can be performed in 1 week using 50–100 GPUs with a computing cost roughly equivalent to the cost for the synthesis of only one new compound. By combining these capabilities with artificial intelligence techniques for goal-directed molecule generation and evaluation, new opportunities can be explored for FEP applications in the drug discovery stages of hit identification, hit-to-lead, and lead optimization based not only on structure exploitation within the given chemical series but also including evaluation and comparison of completely unrelated molecules during structure exploration in a larger chemical space. XFEP provides the basis for scalable FEP applications to become more widely used in drug discovery projects and to speed up the drug discovery process from hit identification to preclinical candidate compound nomination.

Structural insights into the mechanism of nucleotide regulation of pancreatic KATP channel

“…Relative binding free energy (RBFE) calculations are the most common and widespread applications of FEP for drug discovery projects. We used the XFEP platform which have been developed by XtalPi for the FEP calculations 49 The XFEP platform constitutes the backbone of an integrated workflow coupling FEP calculations with AI models and wet-lab experiments. XFEP utilizes the AMBER software package 46 for free energy calculations.…”

“…XFEP utilizes the AMBER software package 46 for free energy calculations. All simulations were performed using the TIP3P water model, as well as AMBER ff14SB for the proteins and XForce Field (XFF) 49 , developed by XtalPi, with a system-specific force field refinement protocol for the ligands. All simulations used a Langevin integrator with a 2 fs timestep for heating and equilibration, 4 fs for production, and a friction coefficient of 2 ps −1 .…”

The long road to calibrated prediction uncertainty in computational chemistry

“…When the uncertainties of the reference data are negligible, this approach tests directly the quality of the expanded uncertainties for the predicted values U (V ) 95 . Otherwise, the quality of the reference set uncertainties U (R) 95 will also affect the results.…”

“…For the reference data, in absence of specific information, I assumed that expanded uncertainties were used (U (R) 95 ). The process of PU estimation in the FPD method has been summarized above (Section II B 1), and it is difficult to infer its nature, beyond a possible over-estimation implied by the worst-case scenario strategy.…”

Structure-based virtual screening workflow to identify antivirals targeting HIV-1 capsid

“…To our knowledge, those two studies are among the first large-scale applications of ABFE in a virtual screening campaign. In recent years automated high-throughput explicit solvent ABFE protocols are being rapidly developed for docking refinement in virtual screening on GPU and cloud computing platforms [ 29 – 31 ]. We also note that similar virtual screening workflows have been successfully applied in recent drug discovery studies against SARS-CoV-2 [ 32 , 33 ], and Hepatitis B Virus (HBV) capsid [ 34 ].…”