What is AMBER?
At its core, AMBER – the acronym for Assisted Model Building with Energy Refinement (AMBER) – is the “. . . collective name for a suite of programs that allow users to carry out molecular dynamics simulations, particularly on biomolecules.” (AMBER 2018 Reference Manual.) It primarily focuses on enabling simulation of enzymes involved in biocatalysis and the interaction between drug candidates and protein receptors. It is a tool broadly used across the pharmaceutical industry to aid in target identification and validation, lead discovery and lead optimization. According to the NCBI National Center for Biotechnology Information (NCBI), “Modeling and simulation techniques have for some time been an important feature of biocatalysis research, often applied as a complement to experimental studies.”
Delving deeper, AMBER consists of both a set of molecular mechanical force fields for the simulation of biomolecules as well as the molecular simulation suite that implements these force fields.
AMBER was established through the foundational contributions of Professor Peter A. Kollman at the University of California, San Francisco in the 1970’s.
Fast forwarding to today, the current AMBER software suite is divided into two parts:
- AmberTools18, a collection of freely available programs for biomolecular simulation and analysis.
- Amber18, which is centered around the pmemd simulation program (“pmemd” stands for “Particle Mesh Ewald Molecular Dynamics”).
Building on a succession of increasingly more functionally-rich and higher performing releases – the previous being AMBER 16 – AMBER 18 brings new functionality, in particular GPU accelerated Thermodynamic Integration (TI) and Free Energy Perturbation (FEP) methods, as well as a 15 to 20% performance boost over AMBER 16 on existing GPUs.
Released on April 17, 2018, AmberTools18 features the following major packages:
- NAB/sff: a program to build molecules, run MD or apply distance geometry restraints using Generalized Born (GB), Poisson-Boltzmann or 3D-RISM implicit solvent models.
- antechamber and MCPB: programs to create force fields for general organic molecules and metal centers.
- tleap and parmed: basic preparatory tools for AMBER simulations.
- sqm: semiempirical and DFTB quantum chemistry program.
- pbsa: performs numerical solutions to Poisson-Boltzmann models.
- 3D-RISM: solves integral equation models for solvation.
- sander: aimed at molecular dynamics simulations including combined Quantum Mechanical/Molecular Mechanical (QM/MM) functionality.
- mdgx: a program for pushing the “boundaries of AMBER MD,” primarily through parameter fitting. Also includes customizable virtual sites and explicit solvent MD capabilities.
- cpptraj and pytraj: tools for analyzing structure and dynamics in trajectories.
- MMPBSA.py and amberlite: energy-based analyses of MD trajectories.
With respect to AMBER18 itself, which is primarily the PMEMD package, major new features include:
- New pmemd.gem code for advanced force fields, including AMOEBA.
- Free energy calculations on GPUs.
- GPU support for 12-6-4 ion potentials.
- Domain decomposition for CPU-parallelism.
- Nudged elastic band calculations for pmemd (CPU and partial GPU implementation).
- Constant redox potential calculations, to supplement constant pH simulations.
- Support and significant performance improvements for the latest Volta and Pascal GPUs from NVIDIA.
AMBER18’s Alignment with Advanced HPC Elite Partners, Intel and NVIDIA
Advanced HPC working directly with key developers of the AMBER pmemd software have developed a number of solutions that offer the best price performance ratio for running AMBER. For example, Advanced HPC is an Elite Intel Partner and by working with a key AMBER developer it was found that the “. . . new Midpoint** based parallel version of pmemd in AMBER18 is now faster and scales better than pmemd from the AMBER 16 version when using Intel© Xeon© Gold 6148 (Skylake) and Intel© Xeon Phi™7250 (Knights Landing) processors.
When AMBER introduced version 11 in 2010, one of the new features at the time was the ability to use GPUs to accelerate pmemd for both explicit solvent particle-mesh Ewald (PME) and implicit solvent Generalized Born (GB) simulations. The capacity of AMBER to utilize GPUs has been significantly broadened in subsequent versions, with the current AMBER 18 demonstrating even measurably better performance and functionality than preceding AMBER iterations.
For example, AMBER18 now supports Thermodynamic Integration (TI) on GPUs as well as the latest NVIDIA GPU components, from the NVIDIA® Tesla® V100 – the world’s most advanced data center GPU ever built to accelerate AI, HPC, and graphics – to the NVIDIA Tesla P4/P40 and critically the extremely cost-=effective GeForce cards such as the GTX1080TI and Titan-XP. Since AMBER is so GPU-intensive, employing the latest NVIDIA GPUs provides the highest caliber of performance and efficiency.
** The Midpoint method for Molecular Dynamics was developed by researchers at IBM as part of their BlueMatter project with BlueGene and subsequently expanded upon and described publicly by researchers at D. E. Shaw Research, LLC.
AMBER 18 Certified Systems
In collaboration with a leading renowned AMBER developer, Advanced HPC now offers fully optimized AMBER 18-centric solutions. These systems come preinstalled with AMBER18 and include 24-hour burn-in with each GPU validated to ensure it provides reproducible and numerically correct results as well as further validate that the performance is at least as good as that reported by the AMBER team on the AMBER website.
We can customize systems to both your anticipated workload and budget. This includes highly cost-effective 4- and 8-way GeForce based systems which offer the highest performance per dollar of any system designed for molecular dynamics.
Why Advanced HPC is Your Go-To AMBER Optimized Solution Provider
- PROJECT MANAGEMENT. Working in conjunction with the aforementioned AMBER specialist, Advanced HPC can handle each and every step of your AMBER project needs, including:
- Custom system design optimized for your anticipated workload and budget.
- System testing, validation and benchmarking.
- Turn-key installation prior to shipping (plus, upgrading and system cleaning when the time comes).
- Coordinating Environment and New Component Install.
- Ongoing interface with the AMBER development team, as necessary.
- STORAGE EXPERTISE. Because Advanced HPC is an industry-leading provider of the latest enterprise storage technology, we can customize a storage infrastructure that not only complements your AMBER solution, but likewise enables its ultimate optimization.
- HANDS-ON CONSULTATION. Advanced HPC system architects and sales engineers are experts in both AMBER and all things HPC and have direct access to key AMBER code authors for technical questions. As such, we can provide a peerless level of AMBER consultation, implementation strategy and ongoing support. (Manage updates and handle any needed patching on an ongoing basis.)
- MUTUAL PARTNERSHIPS. Advanced HPC’s tenured and close working relationships with both Intel (AMBER CPUs) and NVIDIA (AMBER GPUs) will ensure best-in-class performance and system reliability.
- IMMEDIATE PROXIMITY TO A GLOBAL LIFE SCIENCES CAPITAL. The San Diego region is home to more than 1,225 life sciences companies and over 80 independent and university-affiliated research institutes. (Biocom Economic Impact Report, 2017). Advanced HPC is located in the heart of “America’s Finest City,” only minutes away from the 17 million square feet of biopharma lab space, concentrated in mid-county San Diego. We have deep life sciences expertise, experience and relationships that we are eager to leverage on your behalf.
Our team looks forward to working with you to develop a uniquely customized AMBER solution that will squarely meet your needs and requirements. To learn more, please complete the Inquiry Form on this page or contact us directly. In the meantime, the AMBER18 and AmberTools18 Reference Manuals are provided below for complete information.