index.Rmd

---
title       : Reproducible research
subtitle    : Few concepts and a couple of tools, v0.1
author      : Joona Lehtomäki
job         : 
framework   : io2012        # {io2012, html5slides, shower, dzslides, ...}
highlighter : highlight.js  # {highlight.js, prettify, highlight}
hitheme     : tomorrow      # 
widgets     : []            # {mathjax, quiz, bootstrap}
mode        : selfcontained # {standalone, draft}

--- #license

## CC-BY

<img src="assets/img/ccby_badge.png" class="centered-img" title="Dan McGlin, CC-BY 3.0"></img>

--- #attribution

## Attribution

* [Git tutorial](https://github.com/johnmcdonnell/Git-Tutorial) by John McDonnel
* [Introduction to Version Control with a focus on Git](https://github.com/ethanwhite/USU-SWC-Bootcamp/blob/master/version_control_slides.pdf) by Dan McGlinn
* [Reproducible research in computational science](https://github.com/karthikram/smb_git) by Karthik Ram
* [knitr and RStudio](https://github.com/yihui/yihui.github.com/blob/master/slides/2012-knitr-RStudio.html) by Yihui Xie

### Sources for this presentationa available at

* [Reproducible research: Concepts and tools](https://github.com/jlehtoma/reproScience)


--- #content &vcenter

## Content

 1. Background
 2. Concepts
 3. Tools: RStudio and Git

--- #motivation1 bg:url(assets/img/motivation_composite.png) bg-repeat:no-repeat bg-position:center

## My own PhD

--- #open-science &twocol w1:50% w2:50%

## Open science

"Opening up access to the data and software, not just the final publication, is one of goals of the open science movement" <br />(Ram, 2013)

*** left
- Open access
- Open data
- Open notebook science
- Open source software

*** right

<img src="http://upload.wikimedia.org/wikipedia/commons/b/bb/Open_Science_Logo.jpg" class="centered-img" style="max-height:50%;" title="By G.emmerich (Own work) CC-BY-SA-3.0, via Wikimedia Commons"></img>

--- &notitle bg:black

<img src="http://bit.ly/10FmaGf" class="centered-img"></img>

--- #concepts

## Reproducibility

> * With replication, independent investigators address a scientific hypothesis and build up evidence for or against it (Peng, 2011)
> * Allows others to build upon existing work and use it to test new ideas and develop methods (Ram, 2013)
> * While currently there is unilateral emphasis on "first" discoveries, there should be as much emphasis on replication of discoveries ([Ioannidis, 2005](http://www.telegraph.co.uk/technology/3342867/Is-the-spirit-of-Piltdown-man-alive-and-well.html))

--- #problems1 

<img src="assets/img/wolkowich_et_al_data_sharing.png" class="centered-img" title="Wolkowich et al. 2012"></img>

Wolkovich et al. 2012

--- #problems2

<img src="assets/img/peng_comp_repro.png" class="centered-img" title="Peng 2011"></img>

Peng 2011

--- #problems3 bg:url(assets/img/reproducibility_initiative.png) bg-repeat:no-repeat bg-position:center

--- #problems4

<img src="assets/img/sirkia_et_al_workflow.png" class="centered-img" style="margin-left: 0; float: left;" title="Sirkiä et al. 2012"></img>

Sirkiä et al. 2012

--- #problems5

<iframe src="http://bit.ly/10FnSHI"></iframe>

--- #rstudio1

## Tools for reproducible research

<br />

<img src="http://bit.ly/Z2DkiR" class="centered-img" style="max-height:50%;" title="RStudio, GPLv3"></img>

--- #rstudio2 bg:url(assets/img/rstudio_analysis.png) bg-repeat:no-repeat bg-position:center

--- #rstudio3

## Reproducible?

> * Automatically regenerate documents when code, data, or assumptions change
> * Eliminate transposition errors that occur when copying results into documents
> * Preserve contextual narrative about why analysis was performed in a certain fashion
> * Documentation for the analytic and computational processes from which conclusions are drawn

[Xie, 2012](https://github.com/yihui/yihui.github.com)

--- #rstudio4 &notitle

<br />
<br />

<img src="assets/img/tech_logos.png" class="centered-img" style="max-height:100%;" title="RStudio, GPLv3"></img>

--- #rstudio5 bg:url(assets/img/rstudio_knitr_input.png) bg-repeat:no-repeat bg-position:center

--- #rstudio6 bg:url(assets/img/rstudio_knitr_output.png) bg-repeat:no-repeat bg-position:center

--- #rstudio7 bg:url(assets/img/knitr_workflow.png) bg-repeat:no-repeat bg-position:center

<span style="position: fixed; bottom: 75px;">
Adapted from [Markus Kainu](http://markuskainu.fi/r-tutorial/repro/suomeksi.html)
</span>

--- #rstudio8

## More R-code

```{r}
# Quick summary
library(ggplot2)
summary(cars)
```
--- #rstudio8 &notitle

```{r, echo=TRUE,message=FALSE}
# Quick plot of the data
qplot(speed, dist, data=cars) + geom_smooth()
```

--- #rstudio9 &vcenter

[**source**](http://bit.ly/Y8y8uO)


--- #git1

## Tools for reproducible research

<br />

<img src="http://bit.ly/Z2DNBq" class="centered-img" style="max-height:50%;" title="Jason Long, CC-BY 3.0, via Wikimedia Commons"></img>

--- #git2 bg:url(assets/img/github_linux.png) bg-repeat:no-repeat bg-position:center

--- #git3 bg:url(assets/img/github_landing_page.png) bg-repeat:no-repeat bg-position:center

--- #git4

## What is git? #

<br />

![threecommits](assets/img/threecommits.png)

* It stores snapshots of your projects
* ...It also stores the relationships between those snapshots

[McDonnel, 2012](https://github.com/johnmcdonnell/Git-Tutorial)

--- #git5 &image

<img src="http://www.phdcomics.com/comics/archive/phd101212s.gif" class="centered-img"></img>

--- #git6

## Why use git in science?

> 1. Lab notebook
> 2. Facilitating collaboration
> 3. Backup and failsafe against data loss
> 4. Freedom to explore new ideas and methods
> 5. Mechanism to solicit feedback and reviews
> 6. Increase transparency and verifiability
> 7. Managing large data
> 8. Lowering barriers to reuse

--- #git7 &image

<br />
<br />

<img src="assets/img/smb_karihikram_git_workflow.png" class="centered-img"></img>

<br />

Ram, 2013

--- #git8

## Git: steep learning curve?

<img src="http://d22zlbw5ff7yk5.cloudfront.net/images/cm-27811-1508860c9c2366.gif" class="centered-img" style="margin-top: 150px;"></img>

--- #git9

<img src="assets/img/git_cli.png" class="centered-img" title="Nathan de Vries, CC-BY-SA 3.0"></img>

--- #git10

<img src="assets/img/rstudio_git_panel.png" class="centered-img"></img>

--- #git11 bg:url(assets/img/rstudio_git_history.png) bg-repeat:no-repeat bg-position:center

--- #thanks &vcenter

**Thank you!**

--- #reference .reference

### References:

1. Peng RD (2011): Reproducible research in computational science. Science. 334(6060):1226–7. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3383002&tool=pmcentrez&rendertype=abstract  
2. Ram K. (2013): Git can facilitate greater reproducibility and increased transparency in science. Source Code for Biology and Medicine. 8(1):7. Available from: http://www.scfbm.org/content/8/1/7  
3. Sirkiä S, Lehtomäki J, Lindén H, Tomppo E, Moilanen A (2012): Defining spatial priorities for capercaillie Tetrao urogallus lekking landscape conservation in south-central Finland. Wildlife Biology. 2012;18(4):337–53. 
4. Wolkovich EM, Regetz J, O’Connor MI (2012): Advances in global change research require open science by individual researchers. Global Change Biology. 18(7):2102–10. Available from: http://doi.wiley.com/10.1111/j.1365-2486.2012.02693.x