We'll recall it one more time: this is a shared page where workshop participants are invited to contribute with their notes and, even more important, with their agenda for what they want discussed and tried in a given session. Let us know if you want edit access. By attending you are also invited to maintain your own gists so we (and you) can refer to them here. In summary, there will be a public page like this for every Wednesday so anyone case use it as reference, and anyone can find past sessions by content. We can also discuss the possibility of using gitter for C4B related messaging.

Why Serverless Data Science?

So it can travel to where the data is. Taking a clue from the last 10 years:

Science 2009 - Beyond the Data Deluge - Vol. 323, Issue 5919, pp. 1297-1298.

Forbes 2019 - Radical Change Is Coming To Data Science Jobs - March1

What is Serverless anyway?

It is about computing at scale without worrying about managing the servers (yes there are servers, lots of them). And it is also about not having to write code that only works in the server - even that is abstracted by the infrastructre (discuss). It also a misnomer for the most popular execution model of Cloud Computing.

What exactly is running in the server then?

All client side is now JavaScript orchestrating API Ecosystems. Here's a paper discussing those for cancer research: Grossman et. al (2017) Developing Cancer Informatics Applications and Tools Using the NCI Genomic Data Commons API" Cancer Res. 77(21): e15–e18 [PMID29092930]. How much JS could one expect to reuse in the Server-side FaaS? Let's have a look using www.modulecounts.com.

What if you prefer a non Web platform, like RStudio?

• https://shiny.rstudio.com, see how it relies on Plotly for cross-platform visualization ...
• ... oh, look and that, Jupyter uses it too ... and on that note ...

What if you like literate programming, as Jupyter Notebooks?

• from the gods of d3 (document driven documents), see truly portable https://observablehq.com

What if you hate JS?

• What runs under an API is ultimately your choice, so as long as you "eat your own dog food".

Checklist

So, to take full advantage of the global Web Computer, we should now be ready to go over the checklist in the home page:

1. Review mission and approach at cloud4bio.github.io.
2. Make sure to have Chrome browser installed - Chrome's Dev Tools will be our reference IDE for client-side programming.
3. Ask for edit access to this wiki if you plan on attending with some regularity. Note each session is documented in its own page, under the corresponding year in the top right banner of this page.
4. Create a GitHub account if you don't have one already and pick a github client - use github's own if you don't have a favorite already.
5. Get node, Chrome's headless browser, for server-side programming.
6. (Optional) If you're looking for a cool text editor, vscode is a good choice.
7. (Optional) Add bit.ly/cloud4bio to your phone's Home Screen.

Next Meeting Room Schedule (NCI-Shady Grove)

March 6, 2019 -2W914

March 13, 2019 -7E032/34 <-- canceled, NCI's yearly research retreat

March 20, 2019 -TE420

Mar 27, 2019 -2E908

April 3, 2019 -6E032/34

April 10, 2019 -5E030

April 17, 2019 -6E034

April 24, 2019 -5E030

May 1, 8, 15, & 22, 2019 -7E030

May 25, 2019 -4E030

June 5, 2019- Feb 26, 2020 -7E030

References

This we referred to during the session

• Grossman et. al (2017) Developing Cancer Informatics Applications and Tools Using the NCI Genomic Data Commons API" Cancer Res. 77(21): e15–e18 [PMID29092930].
• "Can Programming Be Liberated from the von Neumann Style?" https://www.thocp.net/biographies/papers/backus_turingaward_lecture.pdf
• Alonzo Church https://en.wikipedia.org/wiki/Alonzo_Church
• Lambda calculus https://en.wikipedia.org/wiki/Lambda_calculus
• "How Bright Promise in Cancer Testing Fell Apart" https://www.nytimes.com/2011/07/08/health/research/08genes.html
• "The FAIR Guiding Principles for scientific data management and stewardship" https://www.nature.com/articles/sdata201618
• "The Genome Analysis Toolkit: A MapReduce framework for analyzing next-generation DNA sequencing data" https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2928508/
• Meteor - Build JavaScript Apps http://meteor.com/
• FAIR https://www.nature.com/articles/sdata201618
• Functional https://www.thocp.net/biographies/papers/backus_turingaward_lecture.pdf
• Reproducibility story: https://www.nytimes.com/2011/07/08/health/research/08genes.html
•