[Xenopus] HHMI Janelia - Informatics Jamboree on Embryonic Cell Types in Xenopus (INJECT-X)

[Peshkin, Leonid]

Dear Xenopus community,
     Our sincere apologies if you are receiving this email more than once - we are trying to make sure we reach all interested people.
We (Marc Kirschner and Leon Peshkin) are writing to announce a special workshop, approved and funded by HHMI, and to solicit your input and feedback. Please continue reading even if you are yourself not an expert on any tissue or cell type or think you will not be able to attend.
We would like to hear from you ASAP but ideally before Monday on email to peshkin at gmail.com<mailto:peshkin at gmail.com> with Jamboree in the Subject line, answers to the following questions, which we have to analyze, compile and send back to HHMI.

-------------------------------------------------------  send this back to Leon   --------------------------------------------------------
Questions: (please copy and paste with your answers)
What tissues and cell types should be covered.
Who are the people who need to be there.
Which resources other than listed below we need to include into annotation.
How can we prepare better for the workshop.
Dates which suit you the best among proposed few but also other suggestions.
Skills. What is your level of computer skills - which environments and languages you normally use to look at data.
Any suggestions at all of how to structure the effort.
--------------------------------------------------------------------------------------------------------------------------------------------------

Rationale
The advent of single cell transcriptomics (SCT) has provided developmental biology with a new tool of extraordinary explanatory power. SCT has a special place to play in understanding the origin of cell types, embryonic plasticity, phenotypic states such as proliferation, migration, and competence, and in understanding gene regulatory circuits, and phenotypic variability, with reference to congenital and other forms of disease.  Adequate technology for obtaining single cell data is now available commercially at moderate cost and new methodologies are continually emerging.  There has been a concomitant improvement in visualizing very high dimensional data sets but here there is a lot of room for further developments.  Yet a major gap exists in marrying the new technologies and computational methods with the knowledge base of embryologists, who following years of work have developed an intimate knowledge of the morphogenesis and gene expression of specific tissues, like the vertebrate kidney, the neural crest that gives rise to the entire peripheral nervous system, the somite tissues, or the early multicellular patterning mechanisms of gastrulation, prechordal mesoderm migration, neurulation and endo-mesododerm formation.  For studies of early development through early organogenesis, the frog embryo based on over a century of experimental embryology has arguably made the greatest contributions to our understanding of the origin of the vertebrate body plan and regional early specification.  Today there is no way that any one person can master intimate knowledge of these processes (which more and more incorporates findings from mouse and zebrafish embryology).  Such knowledge is necessary to interpret the highly specific and quantitative gene expression data.  Furthermore virtually no one in the Xenopus field (or other developmental fields) is fully prepared to make use of the evolving complex mathematical tools needed to access the data and to manipulate it.  From the Klein and Kirschner labs we now have an excellent data set identifying 204 annotated embryonic cell types with others yet to be identified.  We propose a hands on workshop to tackle these challenges.
Workshop Logistics
20-25 participants, experts in Xenopus embryology and cell type differentiation getting together to get the data and get trained in using the available tools and bioinformatics methods, basic computer skills needed and most importantly discuss current state of the art in understanding of differentiation and look for new insights.
Funding
Janelia will cover the cost of meals and a room in the guest house for each participant for the duration of the meeting. All expenses are covered except travel. There might be very limited support for travel for people who are essential to the aims of this meeting and are otherwise unable to make it. Please help us identify such people as soon as possible so we can try to help them.
Dates
The workshop will go from Sunday evening (arrival) to Friday afternoon (departure). Currently proposed by Janelia
are three alternative weeks: Nov 12, Dec 10 or Feb 4. (with some openings even later)
Data
We are planning to distribute and use the following datasets:
 -   ~50K single cell profiles across dev stages 8,10,12,14,16,18,20,22  in X. tropicalis.
 -   ~30K single cell profiles across dev stages 9, 12, 17, 23 in X. laevis.
 -   ~14K whole embryo protein abundance profiles for stage V oocyte, egg, stages: 9,12,17,22,24,26,30,42
 -   ~20K phospho peptides for same stages as above
before meeting we might get more stages and cells.
Computers Skills
We do not require any computer skills. Whatever is your level you'll find something to do and learn at the workshop. We plan to take time to teach the basics of command line processing and MATLAB or R or other skills of your choice.  Even if you are limited to using Web front interface, you will learn something and will be able to fully participate.
Draft Aims
Aim 1:  to assemble active researchers in Xenopus embryology and teach
  - the principles of SCT  SPELL OUT,
  - the current powerful methods of mathematical analysis,
  - the computer skills needed to manipulate the data set;

Aim 2: to study the data and assemble an initial draft of
  - the developmental program of the frog identifying lineages, cell types/states,
  - identify new types/states, building on a map constructed by our labs

Aim 3: to produce and publish
  - summaries of the individual efforts during the workshop and technical report   findings of each investigation,
  - data mining protocols accessible to the public on an existing website that   already includes very deep bulk proteomic and transcriptomic data   complementary to the single cell data,
  - facilitate coordinated peer-reviewed publication of the findings in series of   papers that focus on separate developmental questions and cell populations (a special issue of some journal).
Draft Agenda
Day 1:  for experts including interested members of the Janelia community to discuss various general means of analyzing and visualizing high dimensional data obtained from SCT and other experiments.

Day 2: will be to acquaint the participants with the methods available for SCT, the computational skills for the common forms of accessing the data (MATLAB and light UNIX command line scripting), and then the use of the present HMS pipeline leading to highly manipulatable tSNE plots and STRING.

Days 3,4: will be working with the data, which through hands on experience will further familiarize the participants with the computational analysis, including customizing it for purposes that arise.  Participants may want to prearrange consulting with members of their labs or others during the 3 days  to acquire additional  information they need for analysis.  At the end of each day each participant will present a 10 minute summary of findings and questions- which will take 2.5 hrs.

Day 5: and last day will be devoted to longer talks (10 min) and discussion of the writing of the meeting report and plans for further collaboration and follow-up experiments.



-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.mbl.edu/pipermail/xenopus/attachments/20170831/429788c2/attachment-0001.html