Satellite Events

COMBINE (COmputational Modeling in BIology NEtwork)

Organizers: Dagmar Waltemath (Uni Greifswald, DE) & David Nickerson (University of Auckland, NZ).

Summary: COMBINE 2022 will be a face-to-face meeting organised as a satellite to the ICSB 2022 meeting in Berlin. There will be a series of tutorials and breakouts related to various COMBINE standards running from October 6 & 7, followed by talks and posters at the COMBINE Forum on October 8.


Agenda

COMBINE Community meet-up Oct 6-7, 9am-6pm

COMBINE Forum Oct 8, 8am-1:30pm


Location

Humboldt-Universität zu Berlin, Philippstraße 13, Haus 20, 110115 Berlin


Details and call for tutorials and breakouts available at:

https://combine-org.github.io/events/


SUBMISSION: Please use the official ICSB 2022 EasyChair webpage to submit abstracts for talks and posters at the COMBINE Forum:

https://easychair.org/conferences/?conf=icsb2022 system and select "COMBINE: Satellite Event" as your submission topic.



BioModels: Model of the year competition (Luncheon)

Organizers: Rahuman Sheriff (EBI, BioModels Repository, UK)

Summary: BioModels’s Model of the year is a new competition for early career researchers, including but not limited to PhD students and postdocs, to applaud emerging leaders in the area of systems biology modelling. This competition aims to recognize exciting modelling research as well as promote reproducibility and good modelling practice among early career researchers. The competition will be officially launched at ICSB2022. Details on the application process, eligibility criteria, guidance to submit reproducible models (required as part of application) to enter this competition will be discussed during this luncheon workshop. More details are available at https://www.ebi.ac.uk/biomodels/competition/model-of-the-year-2022


Model of the year is scheduled as a luncheon during ICSB.

Date

October 10, 2022

Time

12:30 - 13:30

Location

H4 Hotel, Room Alexander.

Synthetic Biology in the Age of Machine Learning

Organizers: Charlotte Merzbacher, Diego Oyarzún (University of Edinburgh, UK).

Summary: Advances in AI and machine learning promise to revolutionize the design of genetic circuitry in synthetic biology. This satellite event will discuss current work at the intersection of these fields and speculate on future directions. The meeting will be a half-day face-to-face workshop held on Friday, October 7th. There will be talks from selected abstract submissions and a panel discussion, followed by a networking lunch during the main conference. Researchers with relevant work are encouraged to submit following the instructions below.

Details are available at:

https://cjmerzbacher.github.io/icsb-synbio-ml/


Date

October 7, 2022

Time

10:00 - 16:00

Location

Humboldt-Universität zu Berlin, Philippstr. 13, 10115 Berlin, House 18, Room "Maud-Menten".

SUBMISSION: Please use the official ICSB 2022 EasyChair webpage:

https://easychair.org/conferences/?conf=icsb2022 system and select "Synthetic Biology & ML: Satellite Event" as your submission topic.

JIPipe Workshop (Luncheon)

Organizers: Dr. Zoltan Cseresnyes, Ruman Gerst (Uni Jena, DE)

Summary: The continuous development of new microscopy techniques is tracked by the emergence of new or improved image analysis software that can provide the necessary means of quantification. A stalwart amongst these tools is ImageJ, which has served the increasingly demanding needs of the image analysis community for decades. Despite the large community support behind ImageJ that has been keeping the platform up to date, ImageJ is still hindered by the lack of a versatile visual programming language (VPL) that would allow simple creation of fully reproducible and readily expandable workflows of all levels of complexity according to the FAIR (Findability, Accessibility, Interoperability, and Reusability) principles. Here we designed a new VPL, termed Java Image Processing Pipeline (JIPipe, www.JIPipe.org), which provides over 1000 functionalities of ImageJ in a fully visual format. Main features of JIPipe include full reproducibility of the algorithmic procedure, a standardized project- and results format, automatic scalability towards batch analysis, parallel processing, GPU support via CLIJ2, an integration of Cellpose, as well as support for Python and R scripting. JIPipe is fully equipped to be the new tool of choice for both expert and beginner image analysts and can function as a convenient tool where analysis and quantification ideas can be easily cast into visual workflows. These are easy adaptable and adjustable even by experimentalists, thus bridging the gap between image analysts and experimentalists. In this workshop, we will introduce the main features of JIPipe in interactive sessions, where the details of the main graphical user interface will be demonstrated, and a simple but representative image analysis workflow will be built together with the participants. The structure of the individual image operations, as well as of JIPipe’s backbone data structure will be explored with respect to how batch processing and implementation of the FAIR principles are achieved. More complex examples will follow, where the highly sophisticated data annotation system of JIPipe will be introduced via easy-to-follow examples. Projects will include the tracking of nematodes, the quantification of pathogen uptake by immune cells, deep learning-based segmentation tasks, as well as light sheet microscopy data analysis of the kidney.

Date

October 10, 2022

Time

12:30 - 13:30

Location

H4 Hotel, Room Grenander.

Tutorial: Modelling with VCell (Luncheon)

Organizers: Michael L. Blinov and Ion I Moraru (UConn Health)

Summary: VCell (https://vcell.org/) is a comprehensive platform for modeling cell biological systems that is built on a central database and disseminated as a web application. VCell is widely used in dozens of published research projects (https://vcell.org/vcell-published-models), and for teaching at numerous universities. The software is free with automatic installers for Windows, Mac OS and Linux. It provides a biology-based interface for inexperienced modelers and a math interface for math-inclined researchers. VCell platform may run on user’s computer or on provided remote servers enabling running complex simulations from any low-cost laptop. Models can be accessed from anywhere using the VCell database and can be shared among collaborators or made publicly available.


Technical requirements:

Participants are expected to bring their own laptops, wireless internet access is required.

Short workshop description:

We will demonstrate the whole suite of VCell features, including deterministic (compartmental ODE or reaction-diffusion-advection PDE with support for 2D kinematics), stochastic (SSA solvers), spatial stochastic (reaction-diffusion with Smoldyn), hybrid deterministic/stochastic, rule-based (with incorporated BioNetGen software) and network-free agent based simulations, as well as geometries from 2D or 3D microscope images or from idealized analytical expressions and modeling of membrane flux, lateral membrane diffusion and electrophysiology. We will discuss interaction with other tools, such as ImageJ and SBML-supporting simulators, and command-line simulations with runBiosimulations.

The workshop is scheduled as luncheon during the conference.

Date

October 9, 2022

Time

12:30 - 13:30

Location

H4 Hotel, Room Grenander.

Workshop on statistical inference for dynamical systems

Organizers: Daniel Weindl (Helmholtz Center Munich, DE) & Jan Hasenauer (Uni Bonn, DE).

Summary: Mathematical models are essential tools in systems biology and systems medicine. They contribute to an understanding of structure and dynamics of cellular and tissue dynamics far beyond their isolated parts, as well as to the prevention, diagnosis and therapy of diseases. Yet, model topology as well as model parameters are often unknown and need to be inferred from experimental data. Over the last years, there was substantial progress in the development of standardized model selection and parameter estimation pipelines. Furthermore, optimization and sampling methods have been improved, and novel concepts introduced (e.g., at the interface of mechanistic modelling and machine learning). In this workshop, the state-of-the-art in the field of model selection and parameter estimation will be introduced. This will include basics and advanced methods for statistical inference as well as problem standardization such as PEtab. Furthermore, software tools for different programming languages will be presented, along with materials and examples to get started. The workshop is intended for persons interested in parameter estimation and models selection. It will feature introductory level presentations by experts in the field of systems and computational biology. Furthermore, it will provide time and space to work on statistical inference problems together with tool developers.

Date

October 7, 2022

Time

14:00 - 17:30

Location: Humboldt-Universität zu Berlin, Invalidenstraße 42, 10115 Berlin, Mittelbau (in the courtyard), Room 312a, right.

Further information: https://bit.ly/3cDOMAA

Registration: https://bit.ly/3RikfYo


Tutorial: Creating reproducible biochemical modeling workflows in Python

Organizers: Veronica Porubsky, Janis Shin and Lucian Smith (University of Washington, USA).

Summary: The Center for Reproducible Biomedical Modeling will host a tutorial to demonstrate how to build a biochemical model reproducibly, using COMBINE community standards (SBML, SED-ML, OMEX archives), databases (BioModels, SABIO-RK), and Python tools (Tellurium, Antimony, libRoadRunner, PhraSED-ML, SBStoat, and others). Participants will leave with a better understanding of the importance of reproducible methods across a model building and simulation workflow, and will be introduced to resources where they may seek assistance from developers and experts within the COMBINE community to assist with their model building challenges, especially challenges which inhibit reproducibility. Participants will gain practical experience in the import and export of SBML models into standard-compatible tools, data aggregation and curation, writing model equations and defining parameters, annotating models, parameter fitting, writing simulation experiments which follow the MIASE guidelines, simulating models using SED-ML, model verification and validation, and distribution. Participants will have the opportunity to validate that they have independently reproduced the discussed model and simulation results at the end of the tutorial.

Date

October 7, 2022

Time

14:00 - 17:30

Location

Humboldt-Universität zu Berlin, Invalidenstraße 42, 10115 Berlin, Mittelbau (in the courtyard), Room 312, left.


Registration

https://bit.ly/reproducible_workflows_ICSB_2022

Workshop on gene set enrichment (Luncheon)

Organizers: January Mikolaj Weiner (BIH, DE)

Summary: Gene set enrichments (GSEs) remain an important tools to link statistical results of high throughput data sets with biological reality. While the principle is simple, there is a substantial number of variations and special applications, such as combining GSEs with correlation analysis, multivariate modelling or applying to non-standard data sets (such as OLINK, Nanostring, ATAC-Seq or ChIP-Seq as well as metabolic profiling). There is a substantial number of algorithms and packages for GSEs. These are not redundant, but rather suitable for different applications.

In this tutorial, I will show simple yet effective strategies for gene set enrichment analysis in several different scenarios, including metabolomic profiling, multivariate analyses, single cell RNA-Seq, small gene universes, the strengths and weaknesses of different GSE packages and many tips and tricks. To fully take advantage of the workshop, a very basic command of the R programming language is recommended.

This workshop is scheduled as a luncheon during the conference.

Date

October 11, 2022

Time

12:30 - 13:30

Location

H4 Hotel, Room Grenander.