Towards an open data ecosystem for cell migration research

CC BY-SA 4.0 MULTIMOT: towards an open data ecosystem for
cell migration research paola masuzzo @pcmasuzzo computational omics and systems biology group VIB / Ghent University, Ghent, Belgium CC BY-SA 4.0

CC BY-SA 4.0 CC BY-SA 4.0 modified from http://www.nps.gov/crmo/learn/education/activity-5a.htm

CC BY-SA 4.0 An open data ecosystem for cell migration
research can bring enormous benefits • standardization of data and metadata is essential for reproducibility and unambiguous interpretation of experiments • data exchange = faster and easier access to results + quality control + integration with in silico modeling • open data = opportunities for knowledge discovery, data mining, data re-use • formulation of new hypotheses/principles

CC BY-SA 4.0 Local management, analysis and sharing of cell
migration data plan your migration experiment import and store your data explore and analyse your data collective cell migration single-cell migration exchange your data Towards a global repository for cell migration data data and metadata standardization data dissemination CMSO: your input!

CC BY-SA 4.0 A big challenge in cell migration is
data management and dissemination Masuzzo, Trends in Cell Biology, 2015

CC BY-SA 4.0 There are four key elements to software
for cell migration data Guide and capture the experimental set-up Store, structure and query the acquired data (Semi-)automate data exploration, analysis and interpretation Enable data exchange

CC BY-SA 4.0 That’s why we developed CellMissy… Experiment 0
3h 6h wound cells Experiment Manager Data Loader Data Analyzer Masuzzo, Bioinformatics, 2013 - github.com/compomics/cellmissy

CC BY-SA 4.0 CellMissy helps you plan your migration/invasion experiment

CC BY-SA 4.0 Detailed metadata annotation is essential for data
interpretation

CC BY-SA 4.0 Cell migration data are imported automagically Link
to raw images (ordering of imaged wells) Imaging techniques Exposure time, light intensity, magnification, time interval

CC BY-SA 4.0 Structuring the data takes a bit of
effort

CC BY-SA 4.0 Cell migration can occur both in a
collective and in a single-cell fashion Images courtesy of group C. Ampe, UGent, tracked by CELLMIA

CC BY-SA 4.0 This duality in cell migration modes is
reflected in data analysis pipelines Masuzzo, Trends in Cell Biology, 2015

CC BY-SA 4.0 Automated quality control is performed within replicates
… %area increase density KERNEL DENSITY ESTIMATION outlier detection

CC BY-SA 4.0 Automated quality control is performed within replicates
… outlier detection %area increase density KERNEL DENSITY ESTIMATION %area increase density

CC BY-SA 4.0 … as well as across replicates time
(min) Area (μm2) X X X X X X X

CC BY-SA 4.0 Analysis is carried out on conditions after
averaging across the QC replicates time (min) Area (μm2)

CC BY-SA 4.0 Statistical comparison has never been so easy
analysis report with graphs, tables and results velocity (μm2\min) median cell sheet velocity

CC BY-SA 4.0 Single-cell trajectories carry a lot of information!

CC BY-SA 4.0 Cell-centric parameters are computed on each cell
and then averaged for the entire population trajectory 1

and then averaged for the entire population trajectory 1 trajectory 2

and then averaged for the entire population trajectory 1 trajectory 2 trajectory 3

and then averaged for the entire population trajectory 1 trajectory 2 trajectory 3 trajectory 4

and then averaged for the entire population ... trajectory 1 trajectory 2 trajectory 3 trajectory 4 trajectory 5

CC BY-SA 4.0 Step-centric parameters are aggregated values of all
migration steps

CC BY-SA 4.0 Cell-centric parameters are biased because dependent on
cell trajectory length fast moving cells tend to be associated with relatively straight trajectories, high confinement ratios and low turning angles these cells could be seen as a distinct subpopulation if these cells were to be visualized for a longer time, one might actually see that they do not move so fast, nor so straight

CC BY-SA 4.0 Step-centric parameters can overlook the presence of
cell sub-populations density step-centric displacement cell-centric displacement technical replicates BT-549 cancer cells - 37.5 µM Jasplakinolide

CC BY-SA 4.0 Step-centric parameters can overlook the presence of
cell sub-populations MDA-MB-231 cancer cells - 3D ORIS - 0.05µM EGF step-centric turning angle cell-centric turning angle

CC BY-SA 4.0 CellMissy provides an analysis pipeline where both
types of parameters can be computed... acirc 2.66 direct 0.041 x (µm) y (µm) acirc 5.091 direct 0.079

CC BY-SA 4.0 ... together with ‘classical’ cell migration quantities
acirc 2.66 direct 0.041 x (µm) y (µm) acirc 5.091 direct 0.079 MSD (µm²) time-lag time-index dir. ratio

CC BY-SA 4.0 Additional local properties aid the recognition of
different migratory patterns x (µm) y (µm)

CC BY-SA 4.0 These data support analysis on multiple levels

CC BY-SA 4.0 Metadata and/or data can be easily exchanged
between laboratories lab A

between laboratories lab B lab A

between laboratories lab B This is one file in CellMissy! (≈10 MB) lab A

CC BY-SA 4.0 CellMissy provides key functions for the management
and analysis of cell migration data Experimental set-up and annotation Efficient and fully automated data import and storage Data inspection and analysis on multiple levels Tools for metadata/data exchange

CC BY-SA 4.0 Imagine to have a global repository for
cell migration data... Data Repository

CC BY-SA 4.0 Imagine to have a global repository for
cell migration data... Data Repository Local Software

CC BY-SA 4.0 The need of an open data platform
was shared between researchers Cell migration workshop, Ghent, March 2014

CC BY-SA 4.0 The objective of MULTIMOT is to build
an open data ecosystem for cell migration Masuzzo, Trends in Cell Biology, 2014

CC BY-SA 4.0 Such ecosystem requires standardization of the entire
workflow Controlled vocabularies Minimum reporting requirements    Data and metadata standard formats Masuzzo, Trends in Cell Biology, 2014 cell migration dataset

workflow Controlled vocabularies Minimum reporting requirements    Data and metadata standard formats cell migration dataset Masuzzo, Trends in Cell Biology, 2014

workflow Controlled vocabularies Minimum reporting requirements    Data and metadata standard formats Masuzzo, Trends in Cell Biology, 2014 cell migration dataset

CC BY-SA 4.0 Importantly, deposited data are not the end;
an ecosystem enables re-use as well! Masuzzo, Trends in Cell Biology, 2014 standard-compliant cell migration dataset

an ecosystem enables re-use as well! Masuzzo, Trends in Cell Biology, 2014 standard-compliant cell migration dataset Data Repository

an ecosystem enables re-use as well! Masuzzo, Trends in Cell Biology, 2014 standard-compliant cell migration dataset Data Repository global dissemination

an ecosystem enables re-use as well! Masuzzo, Trends in Cell Biology, 2014 standard-compliant cell migration dataset Data Repository global dissemination multiscale and meta-scale analysis algorithms

CC BY-SA 4.0 The integration of data from different study
models would be much easier Masuzzo, Trends in Cell Biology, 2014

CC BY-SA 4.0 Mathematical modeling and numerical simulations complement in
vitro research Masuzzo, Trends in Cell Biology, 2015

CC BY-SA 4.0 The integration of in vitro experiments with
in silico models would benefit cell migration data and new hypotheses qualitative and quantitative modelling translation into a modelling language data analysis and interpretation wet-lab experiments (test of hypotheses)

CC BY-SA 4.0 MULTIMOT pairs wet-lab scientists with relevant BioIT
researchers Participant No Participant organisation name PI Country 1 VIB (VIB) Lennart Martens BE 2 Weizmann Institute of Science (WIS) Benjamin Geiger IL 3 Ghent University (UG) Christophe Ampe BE 4 Radboud UMC (RUMC) Peter Friedl NL 5 Karolinska Institutet (KI) Staffan Strömblad SE 6 The Chancellor, Masters and Scholars of the University of Oxford (UOXF) Susanna-Assunta Sansone UK 7 Dundee University (DU) Jason Swedlow UK 8 London Research Institute (LRI) Erik Sahai UK 9 Universität Duisburg-Essen (UDE) Matthias Gunzer DE 10 SME 1 – Idea Biomedical (IB) Shlomo Turgeman IL 11 SME 2 – MetaViLabs (MVL) David Drell DE capture, dissemination and analysis of multiscale cell migration data for biological and clinical applications

CC BY-SA 4.0 The Cell Migration Standardization Organization needs you!
www.cmso.science CMSO@MULTIMOT

CC BY-SA 4.0 A checklist trial on Minimum Information About
a Cell Migration Experiment (MIACME) • read a cell migration article (materials & methods) • fill in the form reporting relevant information • give a score to each unit of information https://goo.gl/6Yn2xf Image credits: Dr. Aine Prendergast, Systems Biology Ireland, Remedi, NUIG

CC BY-SA 4.0 MULTIMOT will release key solutions for cell
migration ecosystem to take place Community standards for cell migration data Open software to interact with the standards Public cell migration data repository Multiscale, meta-scale analysis algorithms Applicability in defined model systems and personalised healthcare Dissemination and durability

CC BY-SA 4.0 modified from http://www.nps.gov/crmo/learn/education/activity-5a.htm

CC BY-SA 4.0 www.multimot.org www.cmso.science CC BY-SA 4.0

Towards an open data ecosystem for cell migrati...

Towards an open data ecosystem for cell migration research

Other Decks in Science

Featured

Transcript