Schrödinger Suite 2025-3
Version: 2025-3
Installation: /sw/schrodinger/2025-3
Module: schrodinger/2025-3
License: Commercial (Schrödinger License required)
Overview
The Schrödinger Suite is a comprehensive molecular modeling software package for drug discovery and materials science. It provides integrated tools for protein modeling, ligand design, molecular dynamics, quantum chemistry, and virtual screening.
Key Features: - Maestro graphical interface - Desmond molecular dynamics engine - Glide molecular docking - Prime homology modeling and protein structure prediction - Jaguar quantum chemistry - QikProp ADME prediction - FEP+ free energy calculations - BioLuminate antibody modeling - Materials Science Suite
Quick Start
Loading the Module
module load schrodinger/2025-3
Launching Maestro (GUI)
# Standard launch (software OpenGL)
maestro
# Hardware-accelerated graphics (if using VirtualGL)
vmaestro
# Direct command
$SCHRODINGER/maestro
# With specific license server
SCHROD_LICENSE_FILE=27008@license-server maestro
Main Applications
1. Maestro - Graphical Interface
maestro
- Interactive molecular visualization and modeling
- Workflow builder and job management
- Structure preparation and analysis
- Project management
2. Desmond - Molecular Dynamics
$SCHRODINGER/desmond input.msj -HOST localhost:4
- High-performance MD simulations
- GPU-accelerated
- Protein-ligand binding
- Membrane systems
- Free energy calculations
3. Glide - Molecular Docking
$SCHRODINGER/glide input.in -HOST localhost
- High-throughput virtual screening (HTVS)
- Standard precision (SP) docking
- Extra precision (XP) docking
- Induced fit docking (IFD)
4. Prime - Protein Modeling
$SCHRODINGER/prime input.inp -HOST localhost
- Homology modeling
- Loop refinement
- Side chain prediction
- Protein-protein docking
5. Jaguar - Quantum Chemistry
$SCHRODINGER/jaguar run input.in -HOST localhost
- DFT and ab initio calculations
- pKa and tautomer predictions
- Reaction mechanisms
6. QikProp - ADME Prediction
$SCHRODINGER/qikprop input.mae
- Drug-like property prediction
- ADME/Tox screening
- Lipinski rule of five
Common Workflows
Protein Preparation
# Command-line protein preparation
$SCHRODINGER/utilities/prepwizard \
-fillsidechains \
-fillloops \
-propka_pH 7.4 \
-WAIT \
input.pdb output.mae
In Maestro GUI: - Workflows → Protein Preparation Wizard - Preprocess structure - Review and modify structure - Refine structure (optimize H-bonds)
Molecular Docking with Glide
# Step 1: Generate receptor grid
$SCHRODINGER/glide receptor.in -WAIT
# Step 2: Run docking
$SCHRODINGER/glide dock.in -HOST localhost:4 -WAIT
Example grid generation input (receptor.in):
GRID_CENTER 25.0, 30.0, 15.0
INNERBOX 10, 10, 10
OUTERBOX 30, 30, 30
RECEP_FILE protein_prepared.mae
Example docking input (dock.in):
GRIDFILE receptor.zip
LIGANDFILE ligands.mae
PRECISION SP
DOCKING_METHOD confgen
POSES_PER_LIG 5
Molecular Dynamics with Desmond
# Create MD system
$SCHRODINGER/utilities/multisim \
-m input.mae \
-WAIT \
-o output-md.cms \
-set stage.set_family.md.jlaunch.in.file=desmond_md_job.msj
# Run MD simulation
$SCHRODINGER/desmond \
-HOST "localhost:4:gpgpu=1" \
-cfg desmond_md_job.cfg \
-c desmond_md_job.msj \
-WAIT \
output-md.cms \
-mode umbrella
Standard 100 ns MD workflow:
# Using Maestro workflow:
# 1. System Builder → Add membrane/solvent
# 2. Molecular Dynamics → Run Simulation
# 3. Simulation Interactions Diagram → Analyze
Virtual Screening
# High-throughput virtual screening
$SCHRODINGER/vsw \
-SUBHOST localhost:20 \
-glide-ligands library.mae \
-glide-grid receptor.zip \
-glide-precision HTVS \
-glide-docking-method confgen \
-WAIT
Free Energy Calculations (FEP+)
# FEP+ setup and run
$SCHRODINGER/fep_absolute_binding \
protein.mae \
ligands.mae \
-WAIT \
-HOST localhost:4:gpgpu=1
Homology Modeling with Prime
# Build homology model
$SCHRODINGER/prime_automatic_model \
-seq target.fasta \
-template template.pdb \
-o model.mae \
-WAIT
File Formats
Schrödinger uses several proprietary and standard formats:
| Format | Extension | Description |
|---|---|---|
| Maestro | .mae, .maegz |
Native Maestro format (compressed) |
| CMS | .cms |
Chemical system (MD) |
| DMS | .dms |
Desmond molecular system |
| PDB | .pdb |
Protein Data Bank |
| SDF | .sdf, .sd |
Structure Data File |
| MOL2 | .mol2 |
Tripos MOL2 |
| SMILES | .smi |
Simplified molecular input |
| Grid | .zip |
Glide receptor grid |
Job Control and Hosts
HOST Specification
# Local host with 4 CPUs
-HOST localhost:4
# Local host with 4 CPUs and 1 GPU
-HOST "localhost:4:gpgpu=1"
# Multiple hosts
-HOST "localhost:8,compute-node:16"
# Slurm integration (via job scheduler)
-HOST "slurm:4:gpgpu=1"
Job Submission
# Submit job to run in background
$SCHRODINGER/jobcontrol -submit job.sh
# Check job status
$SCHRODINGER/jobcontrol -list
# Stop a job
$SCHRODINGER/jobcontrol -stop job_id
Environment Variables
| Variable | Purpose |
|---|---|
SCHRODINGER |
Installation directory |
SCHROD_LICENSE_FILE |
License server (format: port@server) |
SCHRODINGER_TMPDIR |
Temporary directory for calculations |
SCHRODINGER_HOSTS |
Default host configuration |
Integration with Cluster
Batch Jobs via Slurm
Example Slurm submission script:
#!/bin/bash
#SBATCH --job-name=glide_dock
#SBATCH --cpus-per-task=8
#SBATCH --mem=16G
#SBATCH --time=24:00:00
#SBATCH --partition=general
module load schrodinger/2025-3
# Run Glide docking
$SCHRODINGER/glide dock.in \
-HOST localhost:${SLURM_CPUS_PER_TASK} \
-WAIT \
-NJOBS 1
GPU Jobs
#!/bin/bash
#SBATCH --job-name=desmond_md
#SBATCH --cpus-per-task=4
#SBATCH --gres=gpu:1
#SBATCH --mem=32G
#SBATCH --time=48:00:00
#SBATCH --partition=gpu
module load schrodinger/2025-3
module load cuda/12.5
# Run Desmond MD with GPU acceleration
$SCHRODINGER/desmond \
-HOST "localhost:${SLURM_CPUS_PER_TASK}:gpgpu=1" \
-cfg desmond.cfg \
-c desmond.msj \
-WAIT \
system.cms
Parallel Virtual Screening
#!/bin/bash
#SBATCH --job-name=vsw_screen
#SBATCH --cpus-per-task=32
#SBATCH --mem=64G
#SBATCH --time=72:00:00
module load schrodinger/2025-3
# Virtual screening with 32 parallel subjobs
$SCHRODINGER/vsw \
-SUBHOST localhost:${SLURM_CPUS_PER_TASK} \
-glide-ligands library.maegz \
-glide-grid receptor.zip \
-glide-precision SP \
-WAIT
Python API (Schrödinger Python)
Schrödinger includes its own Python environment:
# Run Python script with Schrödinger modules
$SCHRODINGER/run python my_script.py
# Interactive Python shell
$SCHRODINGER/run python
# Install additional packages
$SCHRODINGER/run pip install package_name
Example Python script using Schrödinger API:
#!/usr/bin/env $SCHRODINGER/run python
from schrodinger.structure import StructureReader, StructureWriter
# Read structures
reader = StructureReader("input.mae")
for st in reader:
# Process structure
print(f"Structure: {st.title}, Atoms: {st.atom_total}")
# Write output
writer = StructureWriter("output.mae")
writer.append(st)
writer.close()
Utilities
Useful command-line utilities:
# Structure format conversion
$SCHRODINGER/utilities/structconvert input.pdb output.mae
# Split multi-structure file
$SCHRODINGER/utilities/maesubset input.mae -n 1-10 -o subset.mae
# Calculate molecular properties
$SCHRODINGER/utilities/ligprep \
-isd input.sdf \
-omae output.mae \
-ph 7.0 \
-pht 2.0 \
-WAIT
# Prepare ligands
$SCHRODINGER/ligprep -isd ligands.sdf -omae ligands_prepared.mae -WAIT
Trajectory Analysis Tools
Schrödinger provides a comprehensive set of utilities for trajectory manipulation and analysis. These tools allow you to process, analyze, and modify molecular dynamics trajectories generated by Desmond.
See: Trajectory Utilities Reference
Common trajectory operations:
- Alignment and centering: trj_align.py, trj_center.py
- Format conversion: trj_convert.py, trj_merge.py, trj_slice.py
- Analysis: trj_cluster.py, trj_essential_dynamics.py, trj_occupancy.py
- Structure manipulation: trj_unwrap.py, trj_wrap.py, trj_parch.py
- Subsystem extraction: trj_extract_subsystem.py
Quick example - align and center trajectory:
# Align trajectory to reference structure
$SCHRODINGER/run trj_align.py system.cms trajectory_trj aligned
# Center solute in simulation box
$SCHRODINGER/run trj_center.py system.cms centered -t trajectory_trj
For detailed documentation on all trajectory utilities, see the Trajectory Utilities Reference.
Documentation
- Local Documentation:
$SCHRODINGER/docs/Documentation.htm - Official Website: https://www.schrodinger.com/
- Support Portal: https://www.schrodinger.com/support
- Release Notes:
$SCHRODINGER/docs/ReleaseNotes.html - Python API:
$SCHRODINGER/docs/python_api/ - Tutorials:
$SCHRODINGER/docs/Tutorials/
Accessing Documentation
# Open main documentation in browser
firefox $SCHRODINGER/docs/Documentation.htm &
# List all documentation
ls $SCHRODINGER/docs/
# Python API documentation
firefox $SCHRODINGER/docs/python_api/index.html &
Tips and Best Practices
- Always Use -WAIT: Include
-WAITflag for command-line jobs to run synchronously - GPU Acceleration: Use Desmond GPU version for MD (10-20x speedup)
- Protein Preparation: Always prepare proteins before docking or MD
- Save Projects: Use Maestro projects (.prj) to organize work
- Checkpoint Files: Desmond creates checkpoints for job recovery
- Memory Management: Large systems may require >32 GB RAM
- Tmp Space: Set SCHRODINGER_TMPDIR to fast local storage for performance
- Licenses: Check license availability before submitting many jobs
Troubleshooting
License Issues
# Check license status
$SCHRODINGER/licadmin STAT
# Test license connection
$SCHRODINGER/licutil -available
# Set license server explicitly
export SCHROD_LICENSE_FILE=27008@license-server
Graphics Problems
# Use software OpenGL (no GPU needed)
maestro -SGL
# Use VirtualGL for hardware acceleration
vmaestro
# Check OpenGL
glxinfo | grep "OpenGL version"
Job Failures
# Check job log files
tail job_name.log
# Monitor job progress
tail -f job_name.log
# Check host configuration
$SCHRODINGER/utilities/jserver -proxy -status
GPU Detection
# Test GPU availability
$SCHRODINGER/run gpu_info.py
# Check CUDA
nvidia-smi
module load cuda/12.5
Memory Issues
- Increase job memory allocation:
-JOBNAME job_name -maxjob 1 -maxsub 1 - Set tmp directory to larger partition:
export SCHRODINGER_TMPDIR=/scratch - Reduce system size or simulation complexity
Common Applications by Research Area
Drug Discovery
- Hit Identification: Glide HTVS, Phase pharmacophore screening
- Lead Optimization: Glide XP docking, Prime MM-GBSA, FEP+
- ADME Prediction: QikProp, pKa predictions
- Binding Analysis: Desmond MD, Simulation Interactions Diagram
Protein Engineering
- Homology Modeling: Prime
- Loop Refinement: Prime loop refinement
- Protein-Protein: Prime PIPER docking
- Antibody Modeling: BioLuminate
Materials Science
- Polymer Modeling: Materials Science Suite
- Surface Interactions: Desmond MD
- Electronic Properties: Jaguar quantum chemistry
Example Workflows
Complete Protein-Ligand Study
# 1. Prepare protein
$SCHRODINGER/utilities/prepwizard \
-fillsidechains -propka_pH 7.4 -WAIT \
protein.pdb protein_prep.mae
# 2. Prepare ligands
$SCHRODINGER/ligprep \
-isd ligands.sdf -omae ligands_prep.mae -WAIT
# 3. Generate receptor grid
$SCHRODINGER/glide receptor.in -WAIT
# 4. Dock ligands
$SCHRODINGER/glide dock.in -HOST localhost:8 -WAIT
# 5. Run MD on top complex
$SCHRODINGER/multisim \
-m top_pose.mae -HOST "localhost:4:gpgpu=1" \
-WAIT -maxjob 0 -o md_system.cms
# 6. Analyze MD trajectory
$SCHRODINGER/run simulations_analysis.py trajectory.xtc
Support
For technical support and questions: - Schrödinger Support: https://www.schrodinger.com/support - Knowledge Base: https://www.schrodinger.com/kb - User Forums: https://www.schrodinger.com/forum - Cluster Admin: Contact cluster administrators for module-specific issues
Related Modules
moe/2024- Alternative molecular modeling suitevmd/2.0- Molecular visualization and trajectory analysisgromacs/2025.0- Open-source molecular dynamicsamber/24- Amber molecular dynamicscuda/12.5- GPU acceleration for Desmond
Sample Data and Tutorials
# Sample data location
ls $SCHRODINGER/samples/
# Tutorial files
ls $SCHRODINGER/docs/Tutorials/
# Example scripts
ls $SCHRODINGER/mmshare-*/python/scripts/
Last Updated: October 2025 Module Maintainer: XLence Cluster Administration