stem cells in their environment
3rd annual systems microscopy consortium meeting, stockholm: extracting sunny features through the swedish summer clouds
06 June 2014
By Davide Danovi
Scandinavia is arguably a 'state of mind', with Stockholm as its capital. To enjoy it with all its contrasts, go there as close to midsummer as possible! Around then, a dip into the cold sea before or after the omnipresent sauna feels like paradise if sampled in the right amount (or as they say over there: 'lagom').
3rd Annual Systems Microscopy NoE consortium venue for the meeting:"Skogshem & Wijk", on Lidingö situated a little bit outside of Stockholm City.
(Image from Systems Microscopy)
A few days ago, on the first day of June, I had the privilege to test this and to participate in the annual Systems Microscopy meeting. This consortium is a life science project spearheading a key enabling methodology based on live cell imaging for the development of next-generation systems biology, and it brings together 17 groups in Europe. Coming from a cell biology background, I tip-toed into what I expected to be a room full of computer geeks, to receive a warm welcome from a friendly community, and to let several ideas for future experiments brew using new tools (ok, and to head-scratch over obscure formulas on the odd slide too).
Jean-Christophe Olivo-Marin from the Institut Pasteur delivered an impressive lecture on "Quantitative Cell Dynamics", describing the "icy platform" and fascinating futuristic tools for image analysis based on the deformable model approach. In simple terms, these are masks around the objects segmented from microscopy images that act like 'mathematical rubber bands' and are deformed by the actual pixels in the image. This allows a much more robust segmentation of imaged cells in real time, for example, as they remain traceable as separated objects even when they appear to touch each other. Without tools like this, it typically proves difficult to segment two very closely spaced cells as two separate objects. The software was developed for 2D time-lapse microscopy and then employed also for 4D approaches (3D in real time). Christoph Sommer from the IMBA Vienna, and developer of "cell cognition", described the pros and cons of supervised and unsupervised machine-learning approaches and how to combine them effectively for image analysis.
Pablo Hernández Varas from the Karolinska Institutet described vinculin-mediated force transmission in adhesion complex regulation. Erik Danen from Leiden University (Netherlands) spoke about integrin signalling in cancer and how modulation of beta-1 or beta-3 integrins can give rise to faster growing or metastasis-prone tumours in experimental models. Also from Leiden, Vasiliki-Maria Rogkoti presented a panel of 46 breast cancer cell lines and its use for the comparison of their gene signatures with their migratory and invasive behaviour. Again from Leiden, Michiel Fokkelman used this panel to set up an RNA-i screen to identify novel candidate metastasis genes. For this, he employed the phagokinetic track (PKT) assay to follow the trails left by cells as they migrate, a bit like the opposite of the Hop-o'-My-Thumb fairytale.
Rafael Carazo Salas from the University of Cambridge (UK) presented his group's impressive work on 3D image-based high-throughput/high-content microscopy pipelines for yeast-based functional genomics studies. Bernd Fischer from the EMBL talked about the ambitious challenge to produce combinatorial RNA-i screens and estimate directed epistatic effects. Also from the EMBL, Julius Hossain showed ways to model mitosis in mammals by registering the shapes and the axes of mitotic spindles from thousands of cells in 4D to analyse conditions that affect cell division.
Uri Alon from the Weizmann Institute (Israel) talked about how his group uses a library of lung cancer cell lines expressing over a thousand uniquely YFP-tagged proteins as a tool to study how the presence and dosage of specific proteins affects the motility of the individual cell within a population. In a much more informal setting, reminding us of his TED talk, Uri described the emotional/theatrical aspects of being a scientist. He ventured in the profound implications of being apparently lost in your own project without a direction, with the inevitable feeling of loneliness we are all familiar with. Tailoring mental maps for this state of mind, his students apparently refer to it as 'being in the cloud'. To which his answer would be 'You must feel miserable'. But also 'Great!' as that is the place where doubting is allowed and the truly innovative ideas take place. His talk echoed deep into the role of science as an open endeavour, and his guitar accompanied by the supporting vocals of the audience ('scooped scooped') reminded everyone of how much we need the help of colleagues and friends to live and love our crazy job.