Computational Robotics

Developed Software:
with links to publications and download pages

OMPL is a lightweight, thread-safe, easy to use, and extensible library for sampling-based motion planning. The code is written in C++, includes Python bindings and is released under the BSD license. OMPL is also integrated with ROS. On top of the OMPL library, we have developed a GUI for rigid body motion planning that allows users to load a variety of mesh formats that define a robot and its environment, define start and goal states, and play around with different planners.
Planner Arena
Planner Arena is a site for benchmarking sampling-based planners. The site is set up to show the performance of implementations of various sampling-based planning algorithms in the Open Motion Planning Library (OMPL).
Robowflex is a library that makes using MoveIt for motion planning easy, providing a high-level C++ API that simplifies many common use-cases in motion planning research, such as benchmarking, customization, and accessing underlying datastructures.
MotionBenchMaker is an extensible, easy-to-use tool to generate and benchmark datasets for manipulation problems. In addition to the tool to generate datasets, MotionBenchMaker comes with a suite of over 40 prefabricated datasets, with 5 different commonly used robots in 8 environments.
The Task-Motion Kit (TMKit) is a framework for Task and Motion Planning. Everyday activities, e.g., setting a table or making coffee, combine discrete decisions about objects and actions with geometric decisions about collision free motion. TMKit jointly reasons about task-level objectives, i.e., choosing actions and objects, and motion-level objectives, i.e., finding collision free paths.
HyperPlan is a hyperparameter optimization framework for selecting, tuning, and optimizing motion planning performance. Given a dataset, the best planner and configuration for that planner can be selected to maximize arbitrary cost functions such as planning time, combined planning time and path cost, etc.
OOPSMP is a package for motion planning that is easy to extend, robust, and efficient. It can be used for motion planning research or as a teaching tool. This package is still available for download, but is no longer further developed (except for minor bug fixes).

Computational Biomedicine

Developed Software:
with links to publications and download pages

APE-Gen is a fast method for generating ensembles of bound pMHC conformations. It generates an ensemble of bound conformations by iterated rounds of (i) anchoring the ends of a given peptide near known pockets in the binding site of the MHC, (ii) sampling peptide backbone conformations with loop modelling, and then (iii) performing energy minimization to fix steric clashes, accumulating conformations at each round.
EnGens is a Python library for anyone interested in studying protein structures and their flexibility. EnGens provides a computational framework for generation and analysis of representative protein conformational ensembles. With EnGens you will be able to process your datasets of both static and dynamic protein structures. EnGens will analyze your dataset and extract a representative ensemble which you can use for downstream tasks such as docking, structural drug design, or other analysis.
A Pipeline for Selection and Structural HLA Modeling of Conserved Peptides from SARS-related Coronaviruses for Novel Vaccine Development.

Improved structure-based peptide-HLA binding affinity prediction.

DINC-COVID, a webserver that accounts for protein flexibility and allows ensemble docking to SARS-CoV-2 proteins using DINC.
A customizable environment for the structural modeling and analysis of peptide-HLA complexes for cancer immunotherapy
DINC is a parallelized meta-docking method for the incremental docking of large ligands (currently using AutoDock Vina)
Metabolite Translator
Metabolite Translator predicts human metabolites for small molecules including drugs. It is build upon a Neural Machine Translation algorithm representing molecules as sequences using the SMILES notation. Metabolite Translator converts the SMILES of the initial molecule into the SMILES representations of the metabolites that can be possibly formed in the human body. The method has been trained on data that cover metabolism of xenobiotics as well as endogenous compounds and therefore it can predict metabolites through a wide range of enzymes including the enzymes of phase I and phase II drug metabolism.
Large-Scale Structure-Based Prediction of Stable Peptide Binding to Class I HLAs Using Random Forests.
A web server for matching 3D structural motifs in proteins against the proteins in the Protein Data Bank. With a plugin for Chimera the match results can be visualized.
FASST Live is a web server for identifying clusters of proteins that share local structural similarity. (Private use)
A web server for reconstructing all-atom protein structures from coarse grain structures. (Private use)