2010 Progress Report: The Texas-Indiana Virtual STAR Center; Data-Generating in vitro and in silico Models of Developmental Toxicity in Embryonic Stem Cells and Zebrafish

EPA Grant Number: R834289
Center: Texas - Indiana Virtual STAR Center
Center Director: Gustafsson , Jan-Ake
Title: The Texas-Indiana Virtual STAR Center; Data-Generating in vitro and in silico Models of Developmental Toxicity in Embryonic Stem Cells and Zebrafish
Investigators: Gustafsson , Jan-Ake , Glazier, James A , Finnell, Richard H.
Institution: University of Houston - University Park , Texas A & M University , Indiana University - Bloomington
Current Institution: University of Houston - University Park , Indiana University - Bloomington , Texas A & M University
EPA Project Officer: Hahn, Intaek
Project Period: November 1, 2009 through October 31, 2012
Project Period Covered by this Report: November 1, 2009 through October 31,2010
Project Amount: $3,190,993
RFA: Computational Toxicology Research Centers: in vitro and in silico Models Of Developmental Toxicity Pathways (2009) RFA Text |  Recipients Lists
Research Category: Computational Toxicology , Human Health , Safer Chemicals


        The overall objectives for the full project are listed below. The scientific objectives for the first year and how they were met are described in the sections for each Investigational area. The outreach and education objectives and results are described under the management section.
Overall Scientific Objectives (3 years):
  1. Generate developmental models suitable for high throughput screening.
A.      Generation of zebrafish developmental models.
B.      Generation of embryonic stem cell (ESC) differentiation models.
C.      Refinement of models for HTS and automation.
D.      Collaboration with international programs for large-scale screening of chemicals.
  1. Generate high information content models on development and differentiation.
A.      Generation of high information content (HIC) zebrafish models.
B.      Generation of HIC ESC models.
C.      Standardization of assays.
D.      Collection of data from development under normal condition.
E.       Collection of data from genetically modified models, which have a specific disrupted   development phenotype.
F.       Collection of data from models treated with a test set of chemicals.
  1. Develop computational models for developmental toxicology with the ultimate aims of first recreating normal development (in wild-type) and then classifying possible mechanisms by which chemical perturbations cause experimentally observed developmental defects.
A.   Generation of a computational model that faithfully recreates the major morphological features of normal wild-type zebrafish development (i.e., segmentation into somites, proper patterning of vascular and neural systems) and the differentiation to three primitive layers (endoderm, mesoderm and ectoderm) in mouse embryonic stem cells.
B.   Once a working model of normal development has been generated, we will carry out a directed series of parameter sweeps to try to create developmental defects in silico.
C.   Comparison of the results of computationally created defects with experimentally generated defects in zebrafish (IA1) and embryonic stem cells (IA2). Best matches between the two datasets will suggest hypotheses about possible mechanisms by which defects occur.
D.   Design of further experiments to test hypotheses generated and validation of them, if possible.
  1. Perform proof-of-concept experiments of the in vitro and in silico test platforms with a blind test of chemicals.
Outreach and Education Objectives (3 years)
  1. Work towards a global harmonization of chemical regulation.

  A. Collaborate with OECD regarding validation and adoption of new methods for the OECD guidelines.

  B. Transfer of TIVS Center results for mathematical calculations of dose-effects regulators in the USA and Europe.

  C. Transfer of TIVS results to the public databases.

  D.  Disseminate results on chemical screening to stakeholders and the general public.

  1. Education of a new generation of multidisciplinary scientists in the fields of chemical screening, computational toxicology and risk assessment

  A.  Three workshops (one/year)

  B.  Co-supervision of students/postdocs

Objectives and starting point of work at beginning of reporting period
Objectives (3 years)
Establish a robust management component to ensure coordination and organization of integrated research activities.
  1. Efficient communication with and timely delivery of reports to the US-EPA/NCER.
  2. Efficient collaboration with other parties in the field of chemical screening.
  3. Continuous information to the TIVS Center partners about project progress and results, as well as quality assurance activities.
  4. Provision of assistance and advice to the TIVS Center partners regarding administrative issues and reporting routines.
  5. Identification of potential obstacles and provision of appropriate solutions.
  6. Review and evaluation of the research performed within the TIVS Center.
  7. Efficient fiscal management of the subcontracts to research sites.
Investigational Area 1: Zebrafish as a model to elucidate the morphological and mechanistic effects of environmental pollutants
1.    Objectives and starting point of work at beginning of reporting period:
        The aim of this study is to develop toxicity assay tools to perform high throughput screening and to produce high information content models. Our overall objectives are to develop an in vivo screening model using the zebrafish embryo as a tool to follow different developmental and patterning processes. The selected endpoints included:
a)      Gastrulation and early embryonic cell movements
b)      Patterning of CNS and neurogenesis
c)       Hematopoiesis and angiogenesis
d)      Yolk utilization and morphological effects on somitogenesis
        Our specific objectives for the first year of the project were to:
  • Advertise, interview and hire postdocs and lab technician.
  • Collect zebrafish screening models already present in the zebrafish research community.
  • Produce new transgenic zebrafish lines.
  • Imaging the normal development of vasculature, somites and axon pathfinding in 2D and 3D.
  • Exchange of data with Partners 2 and 3 to generate in silico models.
  • Generate automation and imaging strategy for toxicity screens using a panel of known chemicals and transgenic fish.
  • Treatment with test chemicals.
      When we wrote the application for this project, University of Houston (UH) had a zebrafish facility. However, during the spring of 2009, the zebrafish facility was closed and decommissioned. When this project was awarded, we built a new zebrafish facility at the Center for Nuclear Receptors and Cell Signaling. The facility opened in December 2009, and is comprised of a 160 square foot room. A new zebrafish facility is under construction, and will be ready during the spring of 2011. This facility of 640 square feet will be equipped with the Techniplast tank system.
Investigational Area 2: The effects of environmental contaminants on mouse embryonic stem cell differentiation
1.    Objectives and starting point of work at beginning of reporting period:
        Our specific objectives for the first year of the project were to:
  • Advertise, interview and hire postdocs and lab technician.
  • Expand ES cell clones from TIGM library.
  • Optimize models of ES cell differentiation.
  • Capture images using newly acquired In Cell technology.
  • Treatment with test chemicals.
        When we wrote the application for this project, the Finnell laboratory was working in Houston at the Texas A&M Institute for Genomic Medicine. Shortly after funding was initiated, Dr. Finnell accepted a position at the University of Texas at Austin, so the lab was shut down and moved from Houston to Austin. This created a time period when there were staffing problems because of the shut down of the laboratory in Houston, and then a protracted waiting period for the Dean of the College of Natural Sciences to approve positions so that postdoctoral fellows and research associates could be recruited and hired at the new facility. With the exception of the In Cell instrumentation, the laboratory now is completely operational and a research associate has been hired and is working on the project. Offers are out to two postdoctoral fellows and we hope to have them in place by February 2011.
        The time line for the objectives was: The IA2 research group will reanimate and perform quality control on 16 ES cell clones within the first 6 months of the project. The culture of these existing cell clones will be performed under TIGM’s SOPs for ES cell production services. Quality controls include quantitative PCR of Neo and Sry alleles to ensure single transgenic inserts and no loss of Y chromosomes. All clones will also undergo positional conformation of transgenic marker via iPCR, and screening for viral, bacterial, and fungal contamination. Reports will be available for all 16 ES clones during the first 6 months of this project. In the event a clone fails QC, an additional clone will be utilized that targets the same gene. In the event that no additional clones are available, the ES cell clone will either be sub- cloned in an attempt to remove mixed or contaminated clones, or an alternative marker will be selected.
        Within the first year of operation a high-throughput system will be established for the 16 ES cell clones plated in six replicates across a 96-well plate. This system will be characterized for expression of each of the 16 marker genes by β-galactosidase staining.
        The techniques utilized for this application are primarily based on standard ES cell expansion protocol used by the IA2 laboratory. These protocols allow for expansion of ES cell clones from single vials to 96- wells, 24-wells, 60mm, or 100mm plates. All expansions utilize standard M15 media and Soriano ES Feeder cells (SNL 76/7 STO). These procedures were established at the beginning of the reporting period.
Investigational Area 3: Development of computer simulations facilitating assessment of toxicity based on perturbed development in zebrafish and mouse embryonic stem cells
1.  Objectives and starting point of work at beginning of reporting period:
Specific Simulation Aim I— Simulate somitogenesis, axonal pathfinding and intersomitic vascularization in normal embryonic zebrafish development.
        Somitogenesis has been modeled in chicken. We are now adapting that model to the zebrafish model to coincide with expected start of data collection from Zebrafish somites in Year 2. Initial rough models of vascularization have been initiated by Abbas Shirinifard. Axonal pathfinding and neuronal development has not yet been begun.
Specific Simulation Aim II— Simulate normal differentiation in cultured murine ESCs.
        Status: Not yet started.
Specific Simulation Aim III— Use the simulations to identify mechanisms that could cause the specific developmental defects identified in known experimental chemical and knock-out screens of zebrafish embryos and murine ESCs. Correlate quantitative and qualitative developmental perturbations with degrees of toxicity for known compounds.
        Baseline data collection was begun by UH collaborators on untreated wildtype fish embryos, as well as initial studies on arsenic treated fish with vascular GFP labels. These initial data sets should serve as a baseline to begin ramp-up in earnest in year 2.
Specific Simulation Aim IV— Implement further in silico experiments to determine which among the multiple possible mechanisms are more or less likely in explaining defects resulting from exposure to unknown agents.
        Will depend on completion of aims I-III.
Specific Simulation Aim V— Document and deploy the simulations from Specific Simulation Aim IV for use by other actors such as the STAR Centers or European research projects.
        Drs. Glazier, Swat, and Sluka’s travels and meeting attendance in 2010 have been geared towards gathering input from diverse members of the modeling and simulations community to ensure that Aim V will be met.


Progress Summary:

Progress towards objectives and results of the first year
Quality management
        A quality management plan (QMP) was written by the TIVS Center personnel and reviewed by QMM and by EPA (Lisa Doucet, NCER). The approved QMP describes quality management (QM) practices for monitoring and measurement activities, collection of toxicology data, statistical analysis and mathematical modeling in the project. The QM practices are specifically designed to generate and process data of known and appropriate quality. This QMP also summarizes the quality assurance structure of the Center, and identifies the quality control responsibilities of all researchers operating under the Center. It is posted at: http://glazier.biocomplexity.indiana.edu/TIVS_Wiki/Adminsitrative_Reporting/Quality_Management
        For the three investigational areas, sub-project specific quality assurance plans are written (QAPPs). They are posted at: http://glazier.biocomplexity.indiana.edu/TIVS_Wiki/Adminsitrative_Reporting/Quality_Management.
        The quality management will be reviewed during a site visit to University of Houston by QMM on April 25-26, 2011.
TIVS Center Integration
        As the TIVS Center is a virtual one, it was recognized that the distance between the research groups could be an obstacle for efficient collaboration. To overcome this, we developed a TIVS wiki management system to post all results, management documents and relevant publications. The TIVS wiki is password protected and located at http://glazier.biocomplexity.indiana.edu/TIVS_Wiki.
        To promote regular communication, we have had the following meetings:
  • Start up meeting November 19-20, 2009 at UH.
  • Computational Toxicology Subcommittee, September 29-30, 2009. Attended by Maria Bondesson, Richard Finnell and James Glazier. Poster Presentation (EPA).
  • Computational Toxicology Centers STAR Progress Review Workshop, October 1, 2009, EPA Attended by Maria Bondesson, Richard Finnell and James Glazier. Oral Presentation (EPA).
  • Screening models meeting at EPA, May 18, 2010.
  • Regular meetings between UH and Texas A&M/University of Texas at Austin.
  • Regular telephone meetings every second week between UH and IU.
        Telephone meetings:
  • February 26, 2010 with all participants, EPA, QMP Manager
  • November 1, 2010 with all participants and EPA
  • September 22, 2010 with UH, IU and research group of Fatima Merchant to discuss imaging of vascular development in zebrafish
  • October 27, 2010 with UH, IU and research group of Fatima Merchant.
Communication with authorities, stakeholders and non-govermental organizations
         James Glazier, Maciej Swat are leaders of the Interagency Modeling and Analysis Group (IMAG) Working Group 4 (participating agencies are NIH,NSF, NASA, DOE, DOD, USDA, FDA, and Canadian MITACS). The most recent meeting was held September 24, 2010 (http://www.imagwiki.org/mediawiki/index.php?title=Working_Group_4#Working_Group_4:_Cell_Level_Modeling).
Integration with ChemScreen and other European initiatives
  • Maria Bondesson is member of Advisory Board of ChemScreen.
  • Bart van der Burg is member of Advisory Board of TIVS Center.
  • Maria Bondesson participated in ChemScreen start up meeting February 1st, 2010, Amsterdam, The Netherlands.
  • Maria Bondesson participated in ChemScreen meeting September 15-16, 2010, Zeist, The Netherlands.
  • Meeting with CASCADE Acert (Ingemar Pongratz) Houston, December 2, 2010.
  • Netherlands Bioinformatics Center, Workshop Modeling Angiogenesis: Joining Cells, Maths and Computers Oct 4-8, 2010 Leiden, The Netherlands (Swat, Glazier).
  • Jan-Ake Gustafsson spoke at the SAFE Consortium Workshop: ‘Dietary Exposure to Endocrine-Active Pesticides' Brussels, Belgium. 26-27 November 2009.
  • Maria Bondesson attended the Crescendo Integrated Project annual meeting May 27-29, 2010, in Munich, Germany, on Nuclear receptors during development and aging.
Biomedical Simulations using CompuCell3D and Systems Biology Workbench - Training Workshop August 2-13th 2010. Organized by IU.
        The CC3D/SBW training workshop teaches a novel, state-of-the-art approach to multi-cell multi-scale modeling using the Glazier-Graner-Hogeweg (GGH) and Reaction Kinetics network models and the CompuCell3D (CC3D) and Systems Biology Markup Language (SBML) simulation environments. During the workshop, we introduced the mathematical foundations of the both models and walked participants through series of examples of increasing biological and mathematical complexity. We taught Python scripting, which allowed users to build highly complex biomedical simulations integrating both toolkits. By the conclusion of the workshop, participants learned the skills to build models of vascularized tumor growth, segmentation in vertebrate embryos or limb bud development. Because multi-cell simulations such as those done with CC3D (with help of SBW) are becoming an integral part of biomedical research especially in predicting the toxicological impact of various compounds on developing tissues, organs or organisms, it is essential that biologists become familiar with computational techniques used to simulate wide variety of phenomena occurring at multiple scales (molecules to tissues, to organs, to organisms). Simulation environments such as CC3D or SBW greatly reduce efforts necessary to setup simulations and because they are developed by dedicated developer teams so they are much more rigorously tested than typical research simulation codes. However, very few problem solving environments support multi- cell modeling. CompuCell3D (CC3D), developed by Dr. James Glazier, Dr. Maciej Swat and Mr. Randy Heiland, is such a framework. It allows easy model definition, provides sophisticated model customization and allows linking of cell behaviors to underlying biochemical networks inside cells using SBW/SBML. Such models can span multiple scales – from intracellular to tissue level to organ. CC3D/SBW provides a versatile Graphical User Interface which makes visualization, post-processing and on-the-fly analysis of models easy and convenient.
        We estimate that the effort needed to build and run biomedical simulation in one of the problem solving environments (e.g. CC3D) is at least an order of magnitude smaller than coding the model in low level language like C or C++.
      Because both SBW and CC3D build models based on specifying cell behaviors the entry barrier for newcomers is quite low and usually within few hours of training most people (whether they have computational/mathematical training or not) are able to run fairly complex biological simulations
Communication with the general public
     The TIVS center wishes to communicate with the general public to spread information on facts about environmental pollution and development of screening models. To start with, we made a home page for the general public with information on the project (http://tivs-center.uh.edu/).
Investigational Area 1: Zebrafish as a model to elucidate the morphological and mechanistic effects of environmental pollutants
Progress towards objectives and results
Screening models
        The main focus of the first year was to develop/collect the transgenic fish that will be used as screening models. Table 1 lists the different fish strains that we proposed to generate/collect, and the status of these fish strains.
        So far we have generated/collected 5 out of 10 suggested zebrafish strains of the proposal. Two more are available in the community, but they have not been requested yet. In addition, we have collected 5 additional transgenic fish strains that are useful for the project, 3 transparent strains and 4 wt strains.
        The strategy for the development of the screening models is to record the timing and spacial expression of the transgene during normal conditions, and then compare to its expression after chemical treatment. So far, the expression of tissue specific GFP in the different transgenic fish has been followed by time lapse, 3D microscopy. For this purpose a Nikon AZ100 microscope was purchased in the fall of 2010. We also acquired images on Olympus Fluoview confocal microscope in collaboration with Dr. Fatima Merchant, UH. Figure 1 shows pictures of the transgenic fish presently at the facility. Confocal fluorescent microscopy of a time lapse caption of Tg (Flk-1)EGFP showing the growth of intersegmental vessels from dorsal aorta to the dorsal side of the fish is shown in figure 2. Birth and migration of neuronal precursors visualized by Tg (Ngn-1)EGFP fish is shown in figure 3.
        Dharma EGFP fish were constructed in the lab using the Tol2 system. A 5 kb fragment of the Dharma promoter was cloned in front of the EGFP gene. The construct was microinjected into zebrafish embryos at the one cell stage. After 4 months, the fish were genotyped by RT-PCR. The Dharma EGFP expression in the so called organizer of early embryos at the sphere through 80% epiboly stage is shown in figure 4. For details on how to clone with the Tol 2 system, please, see our QAPP at http://glazier.biocomplexity.indiana.edu/TIVS_Wiki/Adminsitrative_Reporting/Quality_Management
High information content models
        For high information content models we have started with the vascular system, specifically the development of ISVs. Through regular meetings with the Indiana University group, we have discussed the parameters for imaging that is required for high information content pictures. For details on these parameters, see: http://glazier.biocomplexity.indiana.edu/TIVS_Wiki/PROJECTS/Project_1__Angiogenesis//_Vascular_Development/Image_Acquisition_Guidelines/Detailed_Image_Acquisition_Parameters
        Several confocal imaging attempts have been made. We have imaged ISV development in 3D and time lapse. The minimum distance that we have been able to take pictures with live embryos is currently approximately 1.5 microns (the distance between z-stacks would ideally be 2x pixel size). All images (raw data) are posted on the TIVA wiki at: http://glazier.biocomplexity.indiana.edu/TIVS_Wiki/PROJECTS/Project_1__Angiogenesis//_Vascular_Development
        To measure length and width of sprouting ISVs, strength of GFP expression and speed of the ISV outgrowth in 3D, we are collaborating with Dr. Fatima Merchant’s research group at Computational Sciences at UH. Dr. Merchant’s group develops programs to analyze bioimages, which is an important tool both for the development of quantitative screening models, and to generate data for high information content models. This data is then transferred to Glazier group for simulations of vascular development (see Investigational Area 3 below).
Toxicity testing
        We started to test toxicity of arsenic and sodium arsenite on the vascular development models. It has previously been reported that the lethal dose of arsenic to zebrafish embryos is 5-10 mM depending on exposure time (Li et al. Aquatic Toxicology (2009) 91:229-237). Li et al. reported that embryos exposed to arsenite (2 mM) exhibited morphological abnormalities such as pericardial edema, abnormal dorsal curvature; flat head; RBC accumulation, and reduced mean body length. We also observed these effect, in addition we observed malformed intersegmental blood vessels (ISVs), normally formed but thin ISV/DLAV, blood clot in trunk/tail region, slow/no blood flow, slow heartbeat, heart defect, malformed head vessels, head hemorrhage, and “loopy” caudal vein plexus in the tail region. Figure 5 shows the effects of arsenite exposure to embryos from day 1 to day 3 with 3mM arsenite; pericardial edema, blood accumulation in the pericardial sac, malformed ISVs and caudal vein. Table 2 shows the preliminary results of the phenotypes after arsenic and arsenite exposure, assessed at 72 hpf.
        The Ngn-1 EGFP transgenic fish have been exposed to cadmium, ethanol, and phenol. We started with these compounds as they have been suggested to disrupt neuronal development. The lethal dose of cadmium was 30mM. At exposure to 3 mM cadmium, we did not observe any difference on GFP expression compared to untreated embryos. Ethanol exposed embryos show many defects. At 3% ethanol, many embryos are malformed. The Ngn-GFP positive cells show a migratory defect, which might reflect a neural tube closure defect. Also at 1 and 2% ethanol, the migratory defect is seen, although to a lesser extent (Figure 6).
Deviation from plans, and corrective actions taken or suggested:
Studies of somite development will just start. These fish have not mated until now.
Investigational Area 2: The effects of environmental contaminants on mouse embryonic stem cell differentiation
Progress towards objectives and results
        The initial work in year one has focused on neuronal differentiation of mouse embryonic stem cells. This data is summarized in the figures below.
        Mapt cells were taken off the fibroblast feeder layer and allowed to form EBs for 2 days and then treated with RA 1 nM (Fig. 1), RA 10 nM (Fig. 2), RA 100 nM (Fig. 3), and RA 1 µM (Fig. 4) and incubated for 4 days. Magnification 200X
        These data indicate RA acid induced neuronal differentiation. 1 µM RA was established as the optimal concentration for the induction of neuronal differentiation.
        Mapt cells were stained ImaGene Green (33 µM, for 30 min) to visualize Mapt promoter driven expression of transgenic β-Gal and then stained with DAPI (300 ng/ml) for 10 min (200X).
        Mapt ES cells were stained with β-Gal as described above and allowed to form EBs for 48 h and then RA (1µM) was added and the process was allowed to continue for another 4 days. EBs were stained with DAPI at the end of the experiment (200X).
        Mapt ES cells were allowed to form EBs for 48 h and neuralized with 1 µM RA for 4 days. EBs were fixed in paraformaldhyde, permeabilized, and stained with anti-β-III tubulin ab (early neuronal marker) and a fluorescent second antibody followed by DAPI.
        Mapt ES cells were allowed to form EBs for 48 h and neuralized with 1 µM RA for 4 days. EBs were fixed in paraformaldhyde, permeabilized, and stained with anti-GFAP antibody (astrocyte marker) and a fluorescent second antibody followed by DAPI.
        Mapt ES cells were allowed to form EBs for 48 h and neuralized with 1 µM RA for 4 days. EBs were fixed in paraformaldhyde, permeabilized, and stained with anti-O4 antibody (oligodendrocyte marker) and a fluorescent second antibody followed by DAPI.
        Mapt cells were allowed to form EBs for 48 h and neuralized with 1 µM RA for 4 days. EBs were dissociated and plated in a differentiation medium containing N2 and B27 in dishes coated with polylysine, polyornithine, entactin, laminin, and collagen. Cells were grown for 8 days and light micrographs were taken (200X).
        Fig. 11 A:  Differentiated cells stained with β-III tubulin antibody, O4 antibody, and DAPI.
        Fig. 11 B:  Differentiated cells stained with DAPI
        Fig. 11 C:  Differentiated cells stained with β-III tubulin antibody.
        Fig. 11 D:  Differentiated cells stained with O4 antibody
Testing Environmental Toxicants on Embryonic Stem Cells
        ES Mapt cells were removed from feeder cells and allowed to form EB bodies in low attachment plates. Arsenite was added with RA to induce neural differentiation. Cell proliferation was determined using
CyQUANT Direct Cell Proliferation Assay (Invitrogen).
Deviation from plans, and corrective actions taken or suggested:
        Funding was made available to IA2 in January of 2010. A research scientist was hired January, 2010 and dedicated solely to the EPA STAR grant. It was determined that breaking up the 16 gene clones into small batches would benefit the collaborative nature of this research by allowing better selection of clones, such that genes/markers identified by IA1 or IA3 could be obtained as mouse ES cells and utilized by IA2 for comparative analyses. To this end, Ngn-1 will be obtained from the KOMP repository to allow comparisons with data generated by IA2 and several genes were selected based on data generated by EPA scientists.
        The IA2 research group initially reanimated, expanded and performed quality control on nine ES cell clones, a total of five genes from the original 16 genes described in the grant proposal. This was accomplished during the first 6 months of the project. Quality control assays were performed on all nine clones. Six of the nine passed the initial QC assays, and the remaining clones were subcloned or reselected. All nine clones passed quality control after subcloning and reselection. An additional five clones are presently undergoing reanimation, expansion and quality control testing. These five clones were agreed upon through correspondence with the EPA, based on data presented by Kelly Chandler. An additional clone was also selected from the Knockout Mouse Project (KOMP) repository. This KOMP clone was agreed upon based on preliminary data produced by IA1. This gene (Ngn1) clone has been ordered, and is presently undergoing quality control testing. An additional four genes/clones are to be identified during year 2 through work with the other members of TIVS or through the EPA.
        The relocation of the IA2 laboratory to the University of Texas at Austin during the course of year one of the project resulted in unfortunate delays. Not only did we lose access to specific pieces of equipment for a period of time, but we lacked sufficient personnel to achieve Year I experimental endpoints. Delays in timing and data production were also associated with this move. However, the reestablishment of both TIGM and IA2 (Finnell Group) in College Station, TX and Austin, TX, respectively, is now complete. In this regard, mouse ES cell culturing is operational at both locations. The live cell high-information, high-throughput screen are expected to be initiated during the first quarter of Year 2 at IA2. These screens will run concurrently to the establishment of vascular differentiation protocols, which will also be moved into screening by the third quarter of year 2.
Investigational Area 3: Development of computer simulations facilitating assessment of toxicity based on perturbed development in zebrafish and mouse embryonic stem cells
Progress towards objectives and results
Year 1/Quarter 1: Personnel
        Began to advertise for and to interview candidates for open personnel positions. We found that the specialized nature of the positions made finding suitable candidates difficult and it seemed that many of the most qualified were either reluctant to come to Bloomington or were able to find positions they liked better. TIVS Wiki was set up to facilitate planning, discussion, and data exchange.
Year 1/Quarter 2: Personnel
        The following four candidates were hired from among over 70 qualified applicants. Based on his unique experience in the field of bioinformatics and database mining, Dr. James Sluka was hired at the senior scientist level to assist in building cell behavior ontologies with the express aim of ensuring that software outputs will be in a form that will be shareable with other groups both within the EPA community and other relevant stakeholder agencies. Dr. Sherry Clendenon was hired to assist the project in the capacity of an experimental developmental biologist based on her expertise both in image collection from embryonic zebrafish and analysis of the image data once it has been collected. In this capacity she has helped to establish strong communications between experimental and modeling subgroups. Dr. Srividhya Jeyaraman was hired to serve as a computational biologist because of her previous experience in simulating biological systems and familiarity with somitogenesis. She will be assisting in developing computational models of somitogenesis and neurogenesis. Mr. Alin Colinescu was has been hired to serve as a software developer with broad-ranging duties including expanding the capabilities of the Compucell software, improving user support, and developing training exercises and manuals, and improving methods of making these available via web-based dissemination.
Quarter 2: Scientific
        Improved CC3D Python API in response to requests from EPA modelers. Parallelized CC3D to significantly improve simulation perfomance. Improved CC3D model description format by including many convenience functions that replaced large snippets of code previously needed to accomplish same task as it is done currently with single function call. Initial simulations of somitogenesis in were roughed in by Susan Hester and Julio Belmonte. Initial models of of intersegmental vasculogenesis were begun by Abbas Shirinifard. Initial neural pathfinding models were delayed by two factors: 1) disruption of the Finnell lab by their move to Austin from Houston and, 2) a lack of access to GFP-labeled neurons in zebrafish which were needed as a basis for model-building.
Quarter 3:
        The initial somitogenesis models were refined further. Developed graphical tool for preparing initial cell layout for simulations. Improved PDE solvers to make them more flexible in the context of developmental simulations. Developed prototype simulations of vascular development. The dynamics of molecular species is linked to cell behaviors to produce regular segments as observed in experiment. Validate, document and release extended version of CC3D. (available on the web at http://www.compucell3d.org/SrcBin ).
Quarter 4:
      The first joint user-training workshop was held at the IUB campus in August 2010, with participation from 4 persons new to CompuCell3D from EPA, 2 new users from UH, and 2 new users from IUB.
        Manuscript reporting multi-scale model of somitogenesis in which segmentation clock is fully simulated at the molecular level was submitted to PLoS Computational Biology in October 2010. The simulations from this manuscript will be added to demonstration software package and user documentation following acceptance and publication.
        First tests of intersegmental vasculogenesis models have been run by Abbas Shirinifard.


Journal Articles: 2 Displayed | Download in RIS Format

Other center views: All 12 publications 2 publications in selected types All 2 journal articles
Type Citation Sub Project Document Sources
Journal Article Bondesson M, Gustafsson J-A. Does consuming isoflavones reduce or increase breast cancer risk? Genome Medicine 2010;2(12):90.
abstract available   full text available
R834289 (2010)
  • Full-text from PubMed
  • Abstract from PubMed
  • Associated PubMed link
  • Full-text: Genome Medicine - Full Text - PDF
  • Abstract: Springer-Abstract
  • Other: ResearchGate - Abstract & Full Text - PDF
  • Journal Article Hester SD, Belmonte JM, Gens JS, Clendenon SG, Glazier JA. A multi-cell, multi-scale model of vertebrate segmentation and somite formation. PLoS Computational Biology 2011;7(10):e1002155.
    abstract available   full text available
    R834289 (2010)
  • Full-text from PubMed
  • Abstract from PubMed
  • Associated PubMed link
  • Full-text: PLoS-Full Text - HTML
  • Abstract: PLoS - Abstract
  • Other: PLoS-Full Text - PDF
  • Relevant Websites:

    http://glazier.biocomplexity.indiana.edu/TIVS_Wiki/Administrative_Reporting/Quality_Management Exit
    http://tivs-center.uh.edu Exit
    http://glazier.biocomplexity.indiana.edu/TIVS_Wiki/Administrative_Reporting/Meeting_Minutes Exit
    http://www.imagwiki.org/mediawiki/index.php?title=Working_Group_4#Working_Group_4:_Cell_Level_Modeling Exit

    Progress and Final Reports:

    Original Abstract
  • 2011
  • Final