TICB 1268 No. of Pages 3

Scientific Life Cell biology, beginning largely as micro- detailed physical–chemical mechanisms Interdisciplinary[3_TD$IF] scopic observations, followed[1_TD$IF]the molec- [7]. The data required for these models ular biology revolution, which viewed are now in sight. New gene editing meth- Team Science in genes, cells, and the machinery that ods are providing endogenous expression underlies their activities as molecular sys- of tagged and mutant cells [8], and new Cell Biology tems that could be fully characterized and live-cell imaging methods are promising Rick Horwitz1,* understood using methods of genetics biochemistry in living cells, measuring con- and biochemistry. Viewing the cell as a centrations, dynamics, equilibria, and complex, dynamic molecular composite organization [9]. Similarly, super-resolution The cell is complex. With its multi- brought insights from chemistry and phys- microscopy and cryoEM tomography, tude of components, spatial– ics to bear on biological problems. Just as which allow structure determination and [6_TD$IF] temporal character, and gene the molecular genetic era was codified by organization in situ [3,4], imaging mass expression diversity, it is challeng- the publication of Watson's book, Molec- spectrometry [10], and single-cell and ing to comprehend the cell as an ular Biology of the Gene [1], two decades spatially-resolved genomic approaches integrated system and to develop later[8_TD$IF]the Molecular Biology of the Cell by [11–13], among other image-based tech- models that predict its behaviors. I Alberts, et al. [2] served a similar purpose nologies, all point to a new golden era of suggest an approach to address for cell biology. cell biology where measurements can be this issue, involving system level performed in cells while still preserving the data analysis, large scale team sci- Currently, cell biology stands in an envi- spatial (and often temporal) character of ence, and philanthropy. able position. The Human Genome the cell. Projecti[4_TD$IF]and related activities have pro- duced a list of molecular players, fueling Computational modeling benefits from an the widespread use of genetic tools and iterative cycle of model and experiment, ‘The area of scientific investigation has producing an increasingly meaningful and ideally executed by integrated, interactive been enormously extended.. But the assimilative power of the human intel- useful understanding of cellular processes collaborations or teams. Furthermore, lect is... strictly limited. Hence, it was and their regulation. Dazzling new light some of the more integrative research inevitable that the activity of the indi- and electron microscopic methods are may require large data sets and a focus, fi vidual investigator should be con ned now allowing the elucidation of complex at least initially, on ‘discovery-driven sci- to smaller and smaller section... as a structures in situ [3,4]. A plethora of ence’. These challenges and complex result of this specialization, it is becom- ‘ ’ ing increasingly difficult for even a omics data on gene expression, post- technologies point to a need for interdis- rough general grasp of science as a translational modifications, proteins, ciplinary teams. This way of performing whole, without which the true spirit of metabolomes, and others are revealing research may require new funding mech- ...’ research is inevitably handicapped important complex relationships and con- anisms, since their scope, absence of - Albert Einstein, The World As I See It nections. Finally, the effects of the micro- clear hypotheses, and possibly the need (1949) environment and mutations on these for large, highly integrated teams, does This quote, written late in Einstein's career, processes, particularly in the context of not readily fit into most NIH funding mech- arguably also describes the present state disease, are revealing insights into basic anisms for basic cell biology. How can of cell biology, sitting at a time of greatly cell biology and its misregulation [5,6]. cell biology foster this new research increased understanding and enormous environment? momentum. Research[7_TD$IF] in cell biology is These new approaches and the conse- poised to deepen our understanding of quent rapid increase in knowledge are Philanthropy is one answer and is playing the mechanisms that underlie the plethora generating immense opportunities for indi- an increasing role in supporting science. In of activities comprising the various cellular vidual investigators;[9_TD$IF] but they are also most cases, it reflects the interests and behaviors that dictate cell types and states. revealing the need for theories and quanti- style of the donor. Many philanthropies While opportunity abounds for continued tative computational models that integrate facilitate research activities focused in a great discovery and understanding, the and conceptualize this knowledge and particular area, and they collaborate with increased specialization that will accom- predict cellular behaviors in response to the academic world to help move impor- pany the deepening investigation presents mutation and environmental alterations. tant cutting edge fields forward. Several a consequent challenge in comprehending This requires quantitative data and a spec- successful examples of philanthropic and and integrating this information. In this per- trum of modeling approaches across a academic relationships exist, including the spective, I discuss a possible approach to range of scales - from machine learning, Broad Institute for studying genomics and address this challenge. correlation and systems approaches to the Wyss Institute for investigating

Trends in Cell Biology, Month Year, Vol. xx, No. yy 1 TICB 1268 No. of Pages 3

translational bioengineering. In addition, multiscale modeling approaches. In this different organelles as well as principles the Howard Hughes Medical Institute way the spirit of the Institute is to and algorithms for cell organization. These (HHMI) was founded to support promising empower, rather than compete with indi- correlations will generate hypotheses that investigators in long-term, high-risk/high- vidual investigators. can then be tested, modeled, and further reward research. The HHMI has a large investigated to determine the mechanisms footprint, and it selects and provides major The Institute's first project takes an inte- by which organelles change positions dur- laboratory funding for over 300 academic grative approach by developing dynamic, ing cellular reorganization. In addition, the investigators in a variety of scientific areas visual data on the organization and local- knowledge of organelle positions and across the United States. This funding has ized activities in human induced pluripotent the timing of their local activities will enable greatly facilitated research in costly areas stem cells (hiPSCs) and cardiomyocytes modeling at many scales, from initial cor- such as investigations focused on struc- derived from them. The choice to study relations to physical–chemical mecha- ture or genetic screens, and in specific hiPSCs was made because they are dip- nisms. An ‘animated cell’ (a cellular areas, such as neuroscience through its loid, ‘disease in a dish’ cell and tissue/ ‘Google Earth’) is a key output for the efforts at Janelia Farm. Other funding organ models that possess relatively stable Institute. The large numbers of replicate mechanisms focus on similar efforts as genomes, propagate well, can be induced measurements allow us to develop posi- the HHMI, including those supported by to differentiate into a number of different tional statistics, which can be visualized in Packard, Pew, MacArthur, the Bur- cell types, and can be used to exploit the data-driven animations. Similarly, since roughs–Wellcome Fund, and The Paul vast and rapidly increasing knowledge of only a few colors can be expressed and G. Allen Frontiers Group. In addition to the human genome [14,15]. imaged in a single cell, data from different these trans-institution mechanisms, there experiments will need to be integrated, are a large number of institutes associated The approach is highly integrated. The using computation-derived positional with specific universities, often aimed at immediate goal is to use gene editing to referencing. These outputs as well as advancing a specific mission by comin- develop a set of well characterized hiPSCs those from other modeling activities can gling scientists in a single location. expressing fluorescence tags on proteins be presented as data-driven animations. that identify the locations of major molecu- The newly-formed Allen Institute for Cell lar structures (molecular machines, organ- Is large scale, team, or philanthropically- Science complements these other insti- elles, and regulatory complexes) found in sponsored research the future of cell biol- tutes in style, mission, and approach. most cells. These cells are also being stud- ogy? Will interdisciplinary or large teams The program has aspirational goals and ied from the undifferentiated state to their with focused goals or resource-rich clearly defined milestones. The overarch- differentiation into cardiomyocytes, which centers replace the traditional, individual- ing mission is to understand and predict is a robust, relatively fast, and reproducible initiated research that has served us so cellular behaviors in normal, pathological, transition. The cells and organoids derived well? Likely not, because small focused and regenerative contexts. Its style is to from hiPSCs are being imaged using a research enterprises addressing specific engage in ‘industrial-scale’ basic research semiautomated pipeline consisting of spin- process are the lifeblood of research, that is open and community-empowering, ning disk and super-resolution modalities. effectively crowdsourcing discovery, lead- but cannot be performed easily in a typical The data are produced in the context of ing to deeper understanding, and fueling academic environment. computational models that will predict translation. But questions and issues positions of the major structures and activ- remain that are not easily addressed by However, the Institute is working closely ities, derive correlations from changes in individuals or small collaborations. There- with the cell biology community. Many of positions, and develop mechanisms for fore, I see a mix of approaches and ways the Institute's activities have been their repositioning. The goal is to produce of studying the cell in the future, with informed and shaped by formal and infor- high replicate measurements(hundreds to investigator-initiated research being in mal conversations with members of the thousands), producing the mean and vari- the majority. The need for other cell biology and allied communities.[10_TD$IF] Open ance of the natural variation in the relative approaches, however, is becoming sharing of tools, reagents (cell lines), and positions for the major cellular compo- increasingly important. While investiga- data are intrinsic to the Institute's activi- nents. The measurements will also be tor-driven research has evolved to work ties. Since[2_TD$IF]the goal of the Institute is to made on cells residing in different environ- well to address specific hypotheses and characterize and integrate a range of cel- ments, in response to perturbations, for problems, the evolution of large-scale lular interactions, rather than to deeply example, drugs and mutation, and during team science is still a relatively new investigate a single process, the Institute changes in cell cycle and differentiation into frontier and successful models for its will add a new ‘scale’ or layer to our under- cardiomyocytes. Statistical analyses of effective implementation are still being standing of the cell and facilitate new these data will reveal coupling among developed.

2 Trends in Cell Biology, Month Year, Vol. xx, No. yy TICB 1268 No. of Pages 3

Resources 4. Asano, S. et al. (2016) In situ cryo-electron tomography: a 10. Caprioli, R.M. (2016) Imaging mass spectrometry: molecu- post-reductionist approach to structural biology. J. Mol. lar microscopy for the new age of biology and medicine. i – www.genome.gov/10001772/all-about-the human- Biol. 428, 332–343 Proteomics 16, 1607–1612 genome-project-hgp/ 5. Nelson, C.M. and Bissell, M.J. (2006) Of extracellular 11. Lubeck, E. and Cai, L. (2012) Single-cell systems biology by matrix, scaffolds, and signaling: tissue architecture regu- super-resolution imaging and combinatorial labeling. Nat. 1Allen Institute for Cell Science, Seattle, WA, USA lates development, homeostasis, and cancer. Annu. Rev. Methods 9, 743–748 – Cell. Dev. Biol. 22, 287 309 12. Trapnell, C. (2015) Defining cell types and states with *Correspondence: [email protected] (R. Horwitz). 6. Engler, A.J. et al. (2009) Multiscale modeling of form and single-cell genomics. Genome Res. 25, 1491–1498 function. Science 324, 208–212 http://dx.doi.org/10.1016/j.tcb.2016.07.007 13. Coulon, A. et al. (2013) Eukaryotic transcriptional dynamics: 7. Ideker, T. and Lauffenburger, D. (2003) Building with a from single molecules to cell populations. Nat. Rev. Genet. scaffold: emerging strategies for high- to low-level cellular 14, 572–584 References – modeling. Trends Biotechnol. 21, 255 262 14. Tiscornia, G. et al. (2011) Diseases in a dish: modeling 1. Watson, J.D. (1965) Molecular Biology of the Gene, Benjamin 8. Doudna, J.A. and Charpentier, E. (2014) Genome editing. human genetic disorders using induced pluripotent cells. 2. Alberts, B. et al. (1983) Molecular Biology of the Cell, The new frontier of genome engineering with CRISPR- Nat. Med. 17, 1570–1576 Garland Science Cas9. Science 346, 1258096 15. Clevers, H. (2016) Modeling development and disease with 3. Liu, Z. et al. (2015) Imaging live-cell dynamics and structure 9. Digman, M.A. and Gratton, E. (2012) Scanning image cor- organoids. Cell 165, 1586–1597 – at the single-molecule level. Mol. Cell. 58, 644 659 relation spectroscopy. Bioessays 34, 377–385

Trends in Cell Biology, Month Year, Vol. xx, No. yy 3