ISMCO’21 Tutorials

Tutorial 1: Current methods and open challenges for structural modeling in cancer immunotherapy (3rd edition)

Understanding the molecular triggers to an immune response is essential for both vaccine development and cancer immunotherapy. In this context, a central step is the activation of T-cell lymphocytes by peptides displayed by a special type of receptor known as Human Leukocyte Antigen (HLA). For instance, a tumor-derived peptide such as MAGEA3 can be used in a vaccine to trigger an immune response against melanoma cells, and MAGEA3-specific T-cells can be expanded ex-vivo or genetically engineered to eliminate cancer cells. Approaches like these have presented exciting results in clinical trials, but have also led to lethal off-target reactions in some patients. Molecular mimicry between the peptide-HLA complexes is a driving factor of these side-effects, making structural analysis an important component in the design of safer immunotherapies. Building on the experience obtained in previous ISMCO meetings, the 3rd edition of this tutorial will introduce attendees to the problem of modeling pHLA complexes, discussing available strategies and open challenges in the field. The attendees will have hands-on experience on how to select, model, and analyze peptide targets for personalized cancer immunotherapy treatments. Special attention will be given to HLA-Arena, a new customizable environment that enables structure-based virtual screening of peptide targets.

Intended Audience
Any person interested in the use of structural bioinformatics methods for cancer immunotherapy applications.

Background assumed
No pre-requisites are required to follow this tutorial. The audience is expected to have a basic understanding of structural biology (e.g., protein primary, secondary and tertiary structure). The audience will also be expected to run a Docker container (i.e., they will have to install Docker in their laptops) (https://docs.docker.com/install/).

Objectives
• Provide an overview on the mechanisms underlying immunotherapy treatments, as well as risks and open challenges.
• Provide an overview on available resources to model and analyze peptide-HLA complexes..
• Provide hands-on on training on how to use 2-3 modeling tools, and 2 visualization software.

Duration
Half-day course.

Activities
• Introductory talk 1 (30 min). The use of adaptive immunity mechanisms for cancer treatment.
• Hands-on exercise 1 (40 min). Exploring IEDB resources
• Hands-on exercise 2 (30 min). Visualizing peptide-HLA structures with Chimera.
• Hands-on exercise 3 (20 min). Ensemble generation with HLA-Arena & APE-Gen.
• Short break (20 min).
• Introductory talk 2 (20 min). Computational strategies to model peptide-HLA complexes.
• Hands-on exercise 4 (40 min). Geometry prediction with HLA-Arena & DINC.
• Hands-on exercise 5 (45 min). Virtual screening with HLA-Arena, APE-Gen and MHCflurry.
• Concluding remarks (15 min). Final remarks and discussion on open challenges.

Organizers
Dinler Antunes, Department of Biology and Biochemistry, University of Houston, USA
Mauricio Rigo, Computer Science Department, Rice University, USA
Andre Fonseca, Department of Biology and Biochemistry, University of Houston, USA
Sarah Hall-Swan, Computer Science Department, Rice University, USA
Lydia Kavraki, Computer Science Department, Rice University, USA

Tutorial 2: How can (experimental) data go on tumor growth models?

Our goal in this tutorial is to show an introduction about modeling in the context of cancer, addressing discrete, continuous, and hybrid models. Examples and interpretation of different types of functions are addressed for the description of tumor growth, mortality, and therapies, among other biological phenomena. We also aim to introduce the attendees to the Python programming language, methods of solving ordinary differential equation models, and model calibration via Bayesian methods. The attendees will have hands-on experience on ordinary differential equation solvers and model calibration.

Intended Audience
Any person interested in the development and calibration of tumor models, or people already familiar with tumor models that are interested in learning how to calibrate these models.

Background assumed
No pre-requisites are required to follow this tutorial. We will give an introduction to Python such as everyone can follow the exercises.

Objectives
• Provide an overview on the development of tumor growth models.
• Provide an introduction to Python and how to solve ordinary differential equations.
• Provide hands-on training on how to calibrate tumor growth models.

Duration
Full-day course.

Activities (tentative)
• Development of tumor growth models (theory): an introduction to discrete, continuous, and hybrid models. Our goal is to familiarize the participants with the biological motivation for using different models of tumor growth.
• Introduction to Python + Development of tumor growth models (hands-on): show basic concepts of Python (variables, read/write files, loops, …) and how to solve ordinary differential equation models and visualize the results.
• Calibration of tumor growth models (theory and hands-on): an introduction to concepts related to Bayesian calibration. We will present a brief introduction to model calibration and the Bayesian theory. In the hands-on part, the attendees will generate in silico data using the model implemented in (2) and calibrate the model parameters using the emcee Python library.

Organizers
Ernesto Lima, Center of Computational Oncology, University of Texas at Austin, USA
Emanuelle Paixao, National Laboratory of Scientific Computing (LNCC), Brazil

ISMCO’20 Tutorials

Current methods and open challenges for structural modeling in cancer

Understanding the molecular triggers to an immune response is essential for both vaccine development and cancer immunotherapy. In this context, a central step is the activation of T-cell lymphocytes by peptides displayed by a special type of receptor known as Human Leukocyte Antigen (HLA). For instance, a tumor-derived peptide such as MAGEA3 can be used in a vaccine to trigger an immune response against melanoma cells, and MAGEA3-specific T-cells can be expanded ex-vivo or genetically engineered to eliminate cancer cells. Approaches like these have presented exciting results in clinical trials, but have also led to lethal off-target reactions in some patients. Molecular mimicry between the peptide-HLA complexes was shown to be the key factor determining these side-effects, making structural analysis an important component in the design of safer immunotherapies. Building on the experience obtained during ISMCO’19, the 2nd edition of this tutorial will introduce attendees to the problem of modeling pHLA complexes, discussing available strategies and open challenges in the field. The attendees will have hands-on experience on how to select, model, and analyze peptide targets for personalized cancer immunotherapy treatments. Special attention will be given to HLA-Arena, a new customizable environment that enables structure-based virtual screening of peptide targets.

Intended Audience
Any person interested in the use of structural bioinformatics methods for cancer immunotherapy applications.

Background assumed
No pre-requisites are required to follow this tutorial. The audience is expected to have a basic understanding of structural biology (e.g., protein primary, secondary and tertiary structure). The audience will also be expected to run a Docker container (i.e., they will have to install Docker in their laptops) (https://docs.docker.com/install/).

Objectives
• Provide an overview on the mechanisms underlying immunotherapy treatments, as well as risks and open challenges.
• Provide an overview on available resources to model and analyze peptide-HLA complexes..
• Provide hands-on on training on how to use 2-3 modeling tools, and 2 visualization software.

Duration
Half-day course.

Activities
• Introductory talk 1 (30 min). The use of adaptive immunity mechanisms for cancer treatment.
• Hands-on exercise 1 (40 min). Exploring IEDB resources
• Hands-on exercise 2 (30 min). Visualizing peptide-HLA structures with Chimera.
• Hands-on exercise 3 (20 min). Ensemble generation with HLA-Arena & APE-Gen.
• Short break (20 min).
• Introductory talk 2 (20 min). Computational strategies to model peptide-HLA complexes.
• Hands-on exercise 4 (40 min). Geometry prediction with HLA-Arena & DINC.
• Hands-on exercise 5 (45 min). Virtual screening with HLA-Arena, APE-Gen and MHCflurry.
• Concluding remarks (15 min). Final remarks and discussion on open challenges.

Organizers
Antunes Dinler, Rice University, USA
Hall-Swan Sarah, Rice University, USA
Lydia Kavraki, Rice University, USA
Rigo Mauricio, Pontifical Catholic University of Rio Grande do Sul, Brazil

ISMCO’19 Tutorials

Current methods and open challenges for structural modeling in cancer immunotherapy

Summary

Understanding the molecular triggers to an immune response is essential to fields such as vaccine development and personalized cancer immunotherapy. In this context, a central step is the activation of T-cell lymphocytes by peptides displayed by a special type of receptor known as Major Histocompatibility Complex (MHC). For instance, a tumor-derived peptide such as MAGEA3 can be used in a vaccine to trigger an immune response against melanoma cells, and MAGEA3-specific T- cells can be expanded ex-vivo or genetically engineered to be even more efficient in eliminating cancer cells. Approaches like these have been already tested in clinical trials with amazing results. However, in a few patients the therapeutic T-cells mistakenly recognized an unrelated peptide, expressed by healthy cells, causing lethal off-target reactions. Molecular mimicry was shown to be the key factor determining these side-effects, making structural analyses an important component in the design of safer immunotherapies. In this tutorial we will discuss the technical challenges of modeling pMHC complexes, and the strengths of available computational strategies. The attendees will have hands-on experience on how to select, model, and analyze peptide targets, and will be introduced to some of the open challenges in the field, regarding sampling, scoring, and cross-reactivity prediction.

Intended Audience
Any person interested in the use of structural bioinformatics methods for cancer immunotherapy applications.

Background assumed

No pre-requisites are required to follow this tutorial. The audience is expected to have a basic understanding of structural biology (e.g., protein primary, secondary and tertiary structure). The audience will also be expected to install python packages using miniconda and to run Docker containers (for those willing to install and run the tools in their own laptops). It is recommended to have Docker Engine (Community) installed before the tutorial (https://docs.docker.com/install/).

Objectives
• Provide an overview on the mechanisms underlying immunotherapy treatments, as well as risks and open challenges.
• Provide an overview on available resources to model and analyze pMHC complexes.
• Provide hands-on on training on how to use 2-3 modeling tools, and one visualization software

Duration:
Half-day tutorial.

Organizers:
Antunes Dinler, Rice University, USA
Abella Jayvee, Rice University, USA
Lydia Kavraki, Rice University, USA
Rigo Mauricio, Pontifical Catholic University of Rio Grande do Sul, Brazil