HPSee HPSee
down

HPSee

HPSee HPSee empowers your team with seamless access to high-performance computing.
Orchestrate jobs, upscale your possibilities and optimize your workflows to achieve uncompromising results in a comfortable timeframe. Try it! unparalleled processing

Your Platform for Streamlined Success

Screen faster and smarter

Harness your compute power and conveniently access virtual screening data.

Upscale your possibilities

Perform extensive computations on remote hardware in a scalable fashion.

Optimize your workflows

Produce uncompromising, high-quality results within a reasonable timeframe.

Your Platform for Streamlined Success

Empowering High-Performance Computing

HPSee is your platform for efficient virtual screening workflow execution. It enables the effortless handling of large compound collections in computational drug discovery scenarios, such as docking on a target structure of interest. After setting up their campaign in BioSolveIT applications with ease, users smoothly initiate the calculations on a hardware running HPSee that utilizes the available resources for fast processing. Once the calculations are finished, users can conveniently access the results for further evaluation. No complicated preparation of libraries and data required! The clean administrative dashboard comes with an intuitive management interface. While operating the platform, admins can effortlessly oversee compound libraries and Chemical Spaces, as well as supervise user permissions.

How You Will Profit from HPSee

IT/Admin

Manage your team's access to computing resources in a sophisticated manner. HPSee was developed to be flexible and scalable with the necessary requirements.

Medicinal Chemists

Effortlessly perform molecular docking of large number of molecules, even without extensive technical expertise.

Computational Chemists

Streamline your virtual screening campaigns in an environment and get rid of unnecessary data juggling.

How You Will Profit from HPSee

Hassle-Free Ecosystem

The perks of HPSee make it an attractive and sustainable option for both academic groups and industrial research teams:
  • No extensive preprocessing and splitting of libraries into smaller chunks required.
  • Removes the need of data juggling.
  • Can be set up on Windows and Linux operating systems.
  • Install on your hardware of choice: It can be even a laptop standing somewhere in your office next to you that will be responsible for the calculations!
  • Streamlines projects and makes large-scale computing accessible to the whole team.

Where Limits Vanish

Scalability is an important factor in state-of-the-art computational drug discovery.
Therefore, HPSee was made to seamlessly and flexibly adapt to your team's needs and tasks. Expand your capabilities to match the growing demands of your operations by sustainably scaling up and out with HPSee.
Parallel processing of submitted tasks makes the computations more efficient and provides users with results faster.

With HPSee the challenge of screening hundred thousands of compounds becomes as simple as handling only a couple hundred.

Housing Large-Scale Compound Databases

A remarkably convenient feature of HPSee is the compound collection management system. Here, admins can upload molecule libraries (*.sdf format) and Chemical Spaces (*.space format), which users can select for their tasks.
During the upload, libraries automatically receive supplementary details (number of listed molecules, upload date, creator's identity) and can be annotated with an additional description.

HPSee eliminates the necessity to transfer large volumes of data for individual projects. Coupled with BioSolveIT's drug discovery dashboard SeeSAR, virtual screening runs can be conveniently initiated by selecting the respective library within the GUI. Subsequently, calculations are executed remotely on HPSee, and upon completion, results can be easily downloaded for subsequent visual assessment.
Say goodbye to shuffling gigabytes or even terabytes of data around!

Taking Molecular Docking to the Next Level

Computational methods, particularly large-scale virtual screening, have established their significance within the early stages of small molecule drug discovery. Ensuring universal access to this method can therefore enhance project performance by expediting results and subsequent processing of individual tasks.

With HPSee virtual screening becomes accessible to everybody. Even computational beginners can effortlessly start their campaigns from SeeSAR's graphical interface without any complicated preadjustments.

Orchestrate Your Team

HPSee's admin dashboard is the background engine for the management of the team.
The admin(s) can add new team or group members as users to grant them access to HPSee's features, such as remote docking. Once added, users can submit their tasks for subsequent processing.

HPSee handles the tasks and users are notified once results are available. No additional monitoring required!
Updates

HPSee Version 2.4 — Electra

The HPSee 2.4 update ensures compatibility with SeeSAR 15.1, keeping both applications working together smoothly for your workflows.

Throughout HPSee, the term 'Fragments' has been renamed to 'Synthons' to align the terminology with SeeSAR and better reflect their role in Chemical Space exploration. This release also adds support for the maxNofExtended property, which provides an upper-bound estimate of the maximum number of results that can be generated from a selected synthon during extension steps.

When upgrading to SeeSAR 15.1, please also update HPSee to version 2.4 to maintain compatibility between both applications.

For older versions and the full changelog, please click here.

Help
Deployment Guide (PDF)

FAQ

  • Download SeeSAR 15
    • Begin by downloading the latest version of SeeSAR 15. This software provides the main interface for Chemical Space Docking® (C-S-D), where you’ll manage your compound screenings and interact with the docking tools.
  • Download HPSee 2.3
    • Next, download HPSee 2.3, a companion tool designed for high-performance docking calculations. HPSee handles the heavy lifting for docking operations, allowing you to run large-scale screenings on a dedicated machine or server.
  • Download the CSD Starter Package
    • The C-S-D Starter Pack includes essential resources, including two setup guides and the ROCK-1 Test Space—a sample chemical space for testing C-S-D.
  • Follow the HPSee Deployment Guide
    • Install HPSee 2.3 on your dedicated machine using the detailed steps in the HPSee Deployment Guide. This setup ensures HPSee is ready to handle the docking tasks as you use SeeSAR for screening.
  • Refer to the SeeSAR Beginner Guide
    • Install SeeSAR 15 on your primary workstation and follow the SeeSAR Beginner Guide to get familiar with its interface and workflows. This guide covers the basics of navigating SeeSAR, initiating docking runs, and working with the 3D visualization tools.
  • Watch the Tutorial Video
    • If you need further assistance, the SeeSAR 14 Tutorial Video provides a visual walkthrough of setting up and using C-S-D in SeeSAR, ideal for new users.
  1. Logical CPU: 32 cores
  2. Disk Space: 500 GB to store all generated data
  3. Memory (RAM): 2GB per core (logical CPU)
Update to the latest driver, and check — even if Windows tells you that you are up-to-date. Lenovo and other computers with onboard graphics, please navigate to this link to check if there is a newer driver available for you.
  1. Linux is supported. The use of a pure Linux system is recommended.
  2. Windows is supported for Windows 10 (version 1903, build 18362) and all later versions since Docker requires WSL2. To check your Windows build version:
    1. windows logo key + R
    2. type winver
    3. click 'OK'
  3. MacOS is currently not supported.
An elaborated list of the respective operating system (OS) requirements can be found here.
Furthermore, HPSee uses Docker as container run time. Please check the associated requirements following this link.

Requirements for C-S-D

C-S-D in the GUI and API flavors runs its calculations on HPSee. HPSee works as a set of services: a database, an API and workers that carry out the calculations. The hardware you need depends on how you deploy HPSee: on one machine, on a cluster or on AWS.

Find your HPSee setup

Each setup has its own hardware requirements. Answer up to three questions to find yours.

Question Your answer Your setup
1. Where do you want to run HPSee? On premises Continue with question 2
On AWS AWS
2. How is your on-premises hardware organized? A cluster controlled by a scheduling system, such as Slurm Small or large cluster, run with Apptainer or a hybrid of Docker and Apptainer
Machines dedicated to HPSee Continue with question 3
3. How many machines do you dedicate to HPSee? One machine Single machine, run with Docker Compose or Apptainer
Several machines Small or large cluster, run with Docker Swarm or Apptainer

HPSee hardware requirements per setup

Small and large clusters differ in the number of worker nodes. The worker requirements in the last row apply to every setup.

Setup Single machine Small cluster Large cluster AWS
How HPSee runs One machine used only by HPSee, with all services on it Docker Swarm or Slurm orchestration with a dedicated manager node Docker Swarm or Slurm orchestration with a dedicated manager node Managed cluster
Worker nodes 1 10–20 150 Up to 300
Database Container on the same machine On the manager node On the manager node Amazon Aurora PostgreSQL
Manager node No separate manager node 8 vCPU and 16 GB RAM, plus the database host (see below) 16 vCPU and 32 GB RAM, plus the database host (see below) 16 vCPU and 32 GB RAM, plus the database
Worker, per job 8 vCPU, 8 GB RAM (16 GB recommended) and at least 10 GB of disk space (20 GB NVMe recommended). To run several jobs on one worker node, multiply CPU and RAM by the number of jobs.

Example: sizing a single machine (Compose Mode)
On a single machine, the database, the API and the worker share the hardware, so their requirements add up. This example uses the recommended values where available.

Resource Database API Worker (1 job) Total
CPU 8+ cores 4 vCPU 8 vCPU 20+ vCPU
RAM 32 GB 8 GB 16 GB 56 GB
Database container memory budget 16 GB Not applicable Not applicable 16 GB, included in the RAM total
Disk 500 GB+ NVMe 100 GB+ 20 GB NVMe 620 GB+
Network Not applicable (loopback) 10 GbE if users download whole libraries 1 GbE 1 GbE, or 10 GbE if users download whole libraries

Requirements per service for IT administrators

HPSee runs as separate services. Docker Compose and Docker Swarm setups run the API, the database and at least one worker. Slurm setups also run the EventBroker. In cluster setups, the API and the EventBroker run as one replica each on the manager node, which also hosts the database. Only the number of workers changes with the size of the setup. Each cell lists the minimum first and the recommended value below it.

Database host Single machine Small cluster Large cluster
CPU 8 vCPU
Recommended: 8+ cores		 8 vCPU
Recommended: 16 vCPU		 16 vCPU
Recommended: 32 vCPU
Host RAM 16 GB
Recommended: 32 GB		 32 GB
Recommended: 64 GB		 96 GB
Recommended: 128–192 GB
Container memory budget* 8 GB
Recommended: 16 GB		 24 GB
Recommended: 48 GB		 80 GB
Recommended: 128 GB
Disk 250 GB NVMe
Recommended: 500 GB+		 1 TB NVMe, ≥ 500 MB/s, ≥ 5,000 IOPS
Recommended: 2 TB, ≥ 1 GB/s		 2 TB NVMe, ≥ 1 GB/s, ≥ 20,000 IOPS
Recommended: 4 TB, separate WAL (write-ahead log) volume
Network Not applicable (loopback) 1 GbE
Recommended: 10 GbE		 10 GbE, required

* From HPSee 3.0, the installer asks for the container memory budget.

Resource API EventBroker Worker (per job)
CPU 2 vCPU
Recommended: 4 vCPU		 1 vCPU 8 vCPU
RAM 4 GB
Recommended: 8 GB		 512 MB
Recommended: 1 GB		 8 GB
Recommended: 16 GB
Disk (/datastorage) At least 20 GB
Recommended: 100 GB+		 Image only, about 1 GB
Recommended: 5 GB with log headroom		 1 GB per concurrent job, at least 10 GB
Recommended: 20 GB NVMe
Network 1 GbE
Recommended: 10 GbE if users download whole libraries		 Shared 1 GbE
Recommended: 10 GbE at 100+ nodes

Worker values apply to one job. To run several jobs on one worker node, multiply CPU and RAM by the number of jobs per node. Disk and network requirements stay the same.

Supported operating systems and SeeSAR client requirements

Category Requirement Details
HPSee server: operating systems
Linux Supported Docker Engine or Apptainer, depending on your setup
Windows Supported Windows 10 (version 1903, build 18362) or later, with WSL2 and Docker Desktop
macOS Not currently supported HPSee server installations are not currently supported on macOS
SeeSAR client
Memory (RAM) 16 GB Required for working with C-S-D results in SeeSAR
Operating systems Fully supported Linux, macOS and Windows

Note: All hardware requirements are intended as guidance. Actual needs depend on your workload, for example the number of jobs that run at the same time. Setups on dedicated machines need hardware reserved for HPSee, because HPSee does not throttle itself for other jobs on the same machine. The deployment options available to you depend on your HPSee version. The HPSee Deployment Guide describes the installation for each setup. We refine these requirements continuously as HPSee evolves and as we gain experience from more deployments, so values may be updated over time.

Benchmark: In an internal benchmark on protein kinase A (PKA) in Enamine’s REAL Space (7.6 × 1010 compounds at the time of testing) using SeeSAR 14.1.1 and HPSee 2.1.1, anchoring required 9.3 hours of computation time, followed by extension steps of 30.8 hours (100 fragments) and 9.2 hours (50 fragments), on 9 computers with 16 cores and 64 GB RAM each. Actual runtimes vary with the target, Chemical Space, C-S-D settings and available hardware.

Enabling Streamlined Virtual Screening

As an environment to enable large-scale virtual screenings, HPSee was developed to orchestrate the calculations and handling of results.
Coupled with other BioSolveIT applications, HPSee efficiently handles large data sets and streamlines the process in a sophisticated manner.

Docking — Virtual Screening

HPSee supports large-scale, structure-based virtual screening campaigns.
Molecule libraries can be uploaded to HPSee and then selected in SeeSAR's remote Docking Mode. Once the run is initiated, HPSee performs the pose generation and scoring of the ligand-target complexes with HYDE.
In the last step, HPSee processes the results and makes them available for visual inspection in SeeSAR.

Chemical Space Docking®

HPSee was developed to facilitate Chemical Space Docking® (C-S-D), a novel structure-based virtual screening method that screens ultra-vast Chemical Spaces for the most promising drug candidates.
Users can conveniently start their C-S-D workflow in SeeSAR's visual interface. HPSee then takes over the calculations and the preparation of the results.
The initial anchoring of synthons can be enhanced by using co-crystallized ligands or predicted binding poses from docking studies as templates for pose generation. This approach accelerates calculations while producing poses that are structurally aligned with the template molecule.

Learn more about C-S-D.
Read more about C-S-D here (PDF).
Learn more about C-S-D in SeeSAR 15 Apollo.

Get Your Hands on the High-Performance Computing Engine HPSee

Download