-
Notifications
You must be signed in to change notification settings - Fork 12
/
index.Rmd
264 lines (208 loc) · 11.4 KB
/
index.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
---
title: "Introduction to bioinformatics"
author: "Laurent Gatto"
date: "`r Sys.Date()`"
site: bookdown::bookdown_site
knit: bookdown::preview_chapter
description: "Course material for the Introduction to bioinformatics (WSBIM1207) course at UCLouvain."
output:
msmbstyle::msmb_html_book:
toc: TRUE
toc_depth: 1
split_by: chapter
split_bib: no
css: style.css
link-citations: yes
bibliography: [refs.bib, packages.bib]
---
# Preamble {-}
The [WSBIM1207](https://uclouvain.be/cours-2022-wsbim1207.html) course
is an introduction to bioinformatics (and data science) for biology
and biomedical students. It introduces bioinformatics methodology and
technologies without relying on any prerequisites. The aim of this
course is for students to be in a position to understand important
notions of bioinformatics and tackle simple bioinformatics-related
problems in R, in particular to develop simple R analysis scripts and
reproducible analysis reports to interrogate, visualise and understand
data in a tidy tabular format.
The course is followed by *Bioinformatics*
([WSBIM1322](https://github.com/UCLouvain-CBIO/WSBIM1322)) and *Omics
data analysis*
([WSBIM2122](https://github.com/UCLouvain-CBIO/WSBIM2122)).
It is interesting to start this course by asking the students, who
have likely not yet been exposed to bioinformatics
> What is bioinformatics?
While [Wikipedia
defines](https://en.wikipedia.org/wiki/Bioinformatics) it as
> Bioinformatics is an interdisciplinary field that develops methods
and software tools for understanding biological data.
this course won't try to answer that question by providing an overview
of the many context where methods and software tools are developed in
the frame of biological data. We will focus on the getting hands-on
experience in data manipulation. Hence, the material and how it is
taught will be focused on practice.
## Motivation {-}
Today, it is difficult to overestimate the very broad importance and
impact of *data*. Given the abundance of data around us, and the
sophistication of tools for their analysis and interpretation that are
readily available, data has become a tool of profound social
change. Research in general, and biomedical research in particular, is
at the centre of this evolution. And while bioinformatics has been
playing a central role in bio-medical research for many years now,
bioinformatics skills aren't well integrated in life science
curricula, limiting students in their career prospects and research
horizon [@WilsonSayres:2018]. It is important for young researchers to
acquire quantitative, computational and data skills to address the
challenges that lie ahead.
This first course will focus on the data skills. It will not teach to
use any specific piece of bioinformatics software. Once one has
identified a relevant piece of software, running it shouldn't be a
major problem[^runningsw]. The important part will be to understand
what it does, why it does it and how to assess if the output can be
trusted. The latter requires to explore and understand the data and
the results, i.e. data skills. As described by @Buffalo:2015, equating
learning a piece of bioinformatics software to learning bioinformatics
is like learning pipetting as a means to learn molecular biology.
[^runningsw]: Although, in all fairness, some bioinformatics software
are notoriously difficult to install and run.
Critical thinking is essential in any aspect of research, and
bioinformatics and data analysis doesn't escape that rule. Teaching
critical thinking is however very difficult, and arguably impossible
without first possessing the skills and master the tools to be
critical about data. Rather than teaching a limited set of software
tools, this course aims at teaching a set of hands-on skills and
methodologies that allow to interrogate and visualise data, and hence
be in a position to be critical with respect to new data. In addition,
while specific software become obsolete and are replaced, or are
specific to one specific field of research, critical investigation of
data will always be required and will never be replaced.
We will be using the [R](https://www.R-project.org/) language and
environment [@R] and the [RStudio integrated development
environment](https://www.rstudio.com/products/RStudio/) to acquire
these data skills. Other interactive language such as
[Python](https://www.R-project.org/) and the interactive [jupyer
notebooks](https://jupyter.org/) would also have been a good fit. One
motivation of this choice is the availability of numerous
R/[Bionductor](https://www.bioconductor.org/) packages [@Huber:2015]
for the analysis of high throughput biology data.
Another reason why the focus of the course ought to be on data skills
is that a notable difficulty in modern, multidisciplinary research is
communication. Wetlab biomedical scientists aren't required to become
bioinformaticians, statisticians, programmers, ... to be outstanding
researchers, but they will need to communicate efficiently with these
experts (and vice versa). What most often unites all these experts is
data, and communication around data is critical. The importance of
critical thinking and communication around data becomes more evident
when one realises that, when tracking work in bioinformatics core
facilities, only a minority of projects were purely routine and that
most researchers came to the bioinformatics core seeking customized
analysis, rather than a standardized package [@Chang:2015].
To illustrate the reasons why R in general (and in the case of
biomedical sciences, Bioconductor in particular) are worth learning, I
provide here some examples where these software are used. From a
bioinformatics point of view:
- Single cell transcriptomics data: [Bioconductor workflow for
single-cell RNA sequencing: Normalization, dimensionality reduction,
clustering, and lineage
inference](https://f1000research.com/articles/6-1158/v1)
- Quantitative proteomics data: [A Bioconductor workflow for
processing and analysing spatial proteomics
data](https://f1000research.com/articles/5-2926/v2).
- Epigenetic data: [Differential methylation analysis of reduced
representation bisulfite sequencing experiments using
edgeR](https://f1000research.com/articles/6-2055/v2)
- ...
In addition, here are three short videos (in French, with subtitles in
multiple languages) by PhD students that assist in the teaching of
this course. The videos illustrate real-work applications of some of
the concepts taught in the two bachelor bioinformatics course (this
one and [WSBIM1322](https://github.com/UCLouvain-CBIO/WSBIM1322)):
- [Julie Devis](https://youtu.be/MOa9fCHqKko) is pursuing a PhD in the
Computational Biology and Bioinformatics Unit with Prof Laurent
Gatto. She uses R and Bioconductor to study the methylation in
cancer germ-line genes.
- [Valentine Robaux](https://youtu.be/XUiOQ0LRKxc) is pursuing a PhD
in the Cardiovascular Research Unit with Profs Sandrine Horman and
Christophe Beauloye. She uses R and Bioconductor to investigate
platelet function.
- [Jean Fain](https://youtu.be/XC6HDdPzPac) is pursuing a PhD in the
Epigenetics Unit with Prof Charles De Smet. He uses R and
Bioconductor to study DNA methylation and its role in cancer.
And more generally, at Google, Pfizer, Merck, GSK, Bank of America,
the InterContinental Hotels Group, Shell, ...
- [Data Analysts Captivated by R's
Power](https://www.nytimes.com/2009/01/07/technology/business-computing/07program.html)
In summary, the overall learning objectives of this course are:
- for students to apply and adapt the general data analysis techniques
and principles that are presented to new data and new contexts;
- establish links between different concepts seen in the course such
as, for example, the importance of tidy data in general, how it
applies to dataframes in R, and how it enables reasoning on the
data;
- become autonomous when being presented with new data and be in a
position to explore and understand them.
## References and credits {-}
References are provided throughout the course. Several stand out
however, as they cover large parts of the material or provide
complementary resources.
The material for the first chapters, covering the *Introduction to
data science with R*, was originally based on the [**Data Carpentry**
Ecology
curiculum](https://datacarpentry.org/lessons/#ecology-workshop)
[@DCRecol]. The main dataset by [Blackmore *et al.*
(2017)](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5544260/),
introduced in 2021, was prepared by the [Bioconductor education
group](https://github.com/Bioconductor/bioconductor-teaching/blob/master/README.md)
as part of their lesson development.
General references for this course are *R for Data Science*
[@r4ds:2017] and *Bioinformatics Data Skills* [@Buffalo:2015].
The [RStudio Cheat
Sheets](https://www.rstudio.com/resources/cheatsheets/) are also a
handy resource and readers will be pointed to specific sheets in the
respective chapters.
This course is being taught by Prof Laurent Gatto with invaluable
assistance from Mr Jean Fain (since 2019), Ms Valetine Robaux (since
2020), Ms Julie Devis (since 2022), Dr Axelle Loriot (2018 - 2021) and
Mr Kevin Missault (2018 - 2020) at the Faculty of Pharmacy and
Biomedical Sciences (FASB) at the UCLouvain, Belgium.
## Pre-requisites {-}
There are no programming or technical pre-requisities for this course,
other than basic computer usage, such as general knowledge about files
(binary and text files) and folders and as well as downloading
files. Familiarity with a spreadsheet editor is helpful for the first
chapter.
Software requirements are documented in the *Setup* section below.
## About this course material {-}
This material is written in R markdown [@R-rmarkdown] and compiled as a
book using `knitr` [@R-knitr] `bookdown` [@R-bookdown]. The source
code is publicly available in a Github repository
[https://github.com/uclouvain-cbio/WSBIM1207](https://github.com/uclouvain-cbio/WSBIM1207)
and the compiled material can be read at http://bit.ly/WSBIM1207.
Contributions to this material are welcome. The best way to contribute
or contact the maintainers is by means of pull requests and
[issues](https://github.com/uclouvain-cbio/WSBIM1207/issues). Please
familiarise yourself with the [code of
conduct](https://github.com/UCLouvain-CBIO/WSBIM1207/blob/master/CONDUCT.md). By
participating in this project you agree to abide by its terms.
## Citation {-}
If you use this course, please cite it as
> Laurent Gatto, Kevin Missault, Manon Martin & Axelle Loriot. (2021,
> September 24). UCLouvain-CBIO/WSBIM1207: Introduction to
> bioinformatics (Version
> v2.0.0). Zenodo. DOI: 10.5281/zenodo.5532658
[](https://zenodo.org/badge/latestdoi/147494586)
## License {-}
This material is licensed under the [Creative Commons
Attribution-ShareAlike 4.0
License](https://creativecommons.org/licenses/by-sa/4.0/).
## Setup {-}
For chapter \@ref(sec-dataorg) about *Data organisation with
Spreadsheets*, a spreadsheet programme is necessary.
We will be using the [R environment for statistical
computing](https://www.r-project.org/) as main data science language.
We will also use the
[RStudio](https://www.rstudio.com/products/RStudio/) interface to
interact with R and write scripts and reports. Both R and RStudio are
easy to install and works on all major operating systems.
Once R and RStudio are installed, a set of packages will need to be
installed. See section \@ref(sec-setup2) for details.