-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathdata_management_hard_sciences.html
654 lines (422 loc) · 21.6 KB
/
data_management_hard_sciences.html
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
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
<!DOCTYPE html>
<html lang="" xml:lang="">
<head>
<title>Introduction to data management in hard science</title>
<meta charset="utf-8" />
<meta name="author" content="Damien Belvèze" />
<script src="libs/header-attrs-2.14/header-attrs.js"></script>
<link href="libs/remark-css-0.0.1/default.css" rel="stylesheet" />
<link rel="stylesheet" href="monstyle.css" type="text/css" />
</head>
<body>
<textarea id="source">
class: center, inverse, middle
# Introduction to data management in hard science
### Damien Belvèze
<small> 29th of april 2022 </small>
<!--in accordance with R usage, I did not insert any command to automatically load the packages used (install.packages), I only call these packages. If they are missing in the user's Rmarkdown environment, the user will only be prompted to install them..
-->
---
background-color: #bebebe
class: middle, center
# Research Data: What are we talking about?
### some definitions on data management
???
The speaker may display presentation notes by pressing the p key
---
* **research data**: "recorded objects (figures, texts, images or sounds), which are used as main sources for scientific inquiry and are generally considered by the scientific community as compulsory to valid scientific results" (OECD, 2007)
--
* **Datasets**« Aggregation (...) of raw data or intermediary data with some unity, gathered in order to form a consistent whole » (Gaillard, 2014)
--
* **Open Science** :
> Open Science is the practice of science in such a way that others can collaborate and contribute, where research data, lab notes and other research processes are freely available, under terms that enable reuse, redistribution and reproduction of the research and its underlying data and methods
[Open Science Definition](https://www.fosteropenscience.eu/foster-taxonomy/open-science-definition).
???
- not only for STEM sciences
- should always have been the paradigm of all Science since the beginning of Internet.
* **research data** :
As a result, this term excludes :
- preliminary analyses,
- plans for projects to come,
- peer reviews,
- mails from colleagues.
- physical objects
- training material
- administrative records
research data are part of the broader category of what Bruno Latour calls _inscriptions_ which means they are preliminary versions of knowledge (before publication)
* **Open Science**
** reusability & redistribution : see for instance how it works with code (software four freedoms : right to run (=reuse) a code whatever the purpose and freedom to redistribute copies of that code. Richard M. Stallman)
** reproducibility : see "reproducibility crisis". The Munafo manifesto states that only 50% of published results in Health Science are reproducible.
---
class: left, middle
## Research Data: what are we talking about?
.pull-left[
<img src="images/data_narrative.jpg" height="450" />
]
.pull-right[
* Raw data
* derivative data
* analysed data (or interpreted data)
> Data doesn’t say anything. Humans say things. (Andreo Jones-Rooy)
]
---
class: left, middle
## Data accessibility : an old issue
As of year 1997, pioneers of the web have been wondering how data could be more structured in order to be more usable.
.center[
<img src="images/5-star-steps.png" height="400" />
]
???
* First level : open licence
* Second level : structured data
* Third level : open format
* Fourth level : linked data (with URI)
* Fifth level : linked with other data (by the mean of ontologies)
---
background-color: #fcf6ef
color: white
class: middle, center
# What about you? What kind of data do you produce?
---
class: left, middle
# Data storage, curation and publication: what is at stake?
## Economic aspects
> Research data in digital form is increasingly used in works beyond the project for which it was originally collected, as well as in other research areas and in industry
[OECD report 2007](https://www.oecd.org/fr/science/inno/38500823.pdf)
???
Data production may sometimes cost a lot of money (let's take research conducted in the abyss
as an example). Raw data should be archived with great care. DNA sequencing is also very expensive. One can understand why universities or laboratories consider that their ROI will not be raised through freely sharing such data. But Science should be more open, since for a great part it is sustained by public funds.
On the other part, the research funders insist on the growth lever that an improved access to research data would create (inspired by the neo-liberal concept of the _knowledge economy_).
---
class: middle, left
## Health issues
### Sharing medical data helps cure people
collecting individual cases data for future meta-analyses has become strategic for public health
>no other patient was recruited for a study ; individual cases data, once gathered, were enough for the Gustave-Roussy Institute to make a meta-analysis that showed that concomitant chemotherapy is more efficient than sequential chemotherapy (Florian Naudet, Claude Pellen)
Source : [The conversation](https://theconversation-com.cdn.ampproject.org/c/s/theconversation.com/amp/quand-partager-les-donnees-issues-des-essais-cliniques-permet-de-mieux-soigner-les-patients-171130)
???
Original quote in French:
>Sans recruter un patient de plus, grâce à une méta-analyse sur données individuelles, l’Institut Gustave-Roussy a réussi à montrer que la chimiothérapie concomitante est plus efficace que le schéma séquentiel (Florian Naudet, Claude Pellen)
---
class: middle, center
## Democracy at stake
### data as a Common
Obépine (research project) = measures the prevalence of Sars-Cov-2 in French waste waters - Request for access to data produced by a public entity
The answer refers to the "Law for a Digital Republic" aka "Lemaire's Law" (2016) whose goal is to improve circulation of data and knowledge from public institutions to the public.
[demande Dada](https://madada.fr/demande/donnees_relatives_a_la_concentra#outgoing-411)
### _data_ or _capta_ ?
personal consent (GDPR)
If you use conversation transcripts or forms, do not forget to collect and confidentially store the consent forms for each transaction (form submission, interview)
???
Common in the sense of Elinor Ostrom theory on Commons
anytime you deal with personal data, you have to submit a declaration ot treatment to your Digitial Protection Officer
---
class: left, middle
# Issues related to data retention and sharing
## Scientific Issues
Science **reproducibility** and **reliability** are at stake
ie. [LancetGate](https://retractionwatch.com/2020/06/04/lancet-retracts-controversial-hydroxychloroquine-study/) The Lancet retracts a paper on HCQ based on data that Surgisphere refused to provide to reviewers.
Lancet stated on June 5th 2020 that...
> Our independent peer reviewers informed us that Surgisphere would not transfer the full dataset [...] As such, our reviewers were not able to conduct an independent and private peer review
[Le Monde, 4 juin 2020](https://www.lemonde.fr/sciences/article/2020/06/04/hydroxychloroquine-trois-auteurs-de-l-etude-du-lancet-se-retractent_6041803_1650684.html)
---
class: middle, left
## Accessing the data is a condition to reproducibility
.pull-left[
<img src="images/reproducible.png" width="900" />
]
.pull-right[
poor quality control on data
Manifesto for reproducible science
]
---
class: left, middle
## ephemeral data
.pull-left[
<img src="images/perte_donnees.png" width="600" />
]
.pull-right[
80% of the data produced these last twenty years are lost.
]
---
class: middle, left
## How it should be and how it is
.pull-left[
<img src="images/data_pyramid1.png" width="600" />
]
.pull-right[
<img src="images/data_pyramid2.png" width="600" />
]
(Marie Puren, 2021)
???
* More than 80% of the data produced is stored elsewhere than in data repositories.
* In a european survey, 90% of the interviewed researchers say that they have to rely on themselves to store, archive and share their data.
Source: [DATACC, Gestion des données, une nouvelle exigence, de nouvelles compétences, 2020](https://www.datacc.org/bonnes-pratiques/adopter-un-plan-de-gestion-des-donnees/gestion-des-donnees-une-nouvelle-exigence-de-nouvelles-competences/)
---
class: middle, left
## National Plan for Open Science (S Plan)
.pull-left[
<img src="images/PNSO.png" width="800" />
]
.pull-right[
* make open access by default for publications
* **Plan S strongly encourages that research data and other research outputs are made as open as possible and as closed as necessary**
* code that underlies the publication should be made available in external repositories
[Plan S: principles and implementations](https://www.coalition-s.org/addendum-to-the-coalition-s-guidance-on-the-implementation-of-plan-s/principles-and-implementation)
]
---
background-color: #bebebe
class: middle, center
# FAIR principles
## Findable / Accessible / Interoperable / Reusable
<img src="images/FAIR2.png" width="500" />
---
class: left, middle
## How to make your data _findable_?
* dataset -> DOI (exemple : [https://doi.org/10.5281/zenodo.5903186](https://doi.org/10.5281/zenodo.5903186))
* data repositories should be well indexed by search engines like [Google data set search](https://datasetsearch.research.google.com/) (SKOS implementation improves this referencing)
---
class: middle, left
## How to make my data accessible ?
> as open as possible, as close as necessary
Be explicit on how the dataset could be reused and shared with others. Choose a licence adapted to these conditions
???
* If the data are made open : choose a creative commons licence
* when data are accessible on demand : explain how to make a request and whom this request should be sent to.
Data that should not be shared
* secret defense
* government proceedings
* corporate secrecy
data that can be shared after anonymisation / pseudonymisation
* personnal data (included sensitive data)
---
class: left, middle
# Citation advantage
> We also find an association between articles that include statements that link to data in a repository and **up to 25.36% (± 1.07%) higher citation impact** on average (Colavizza, Hrynaszkiewicz, 2020)
<img src="images/data_citation_advantage.png" width="800" />
---
class: left, middle
## How to make my data interoperable?
* Use open formats for your data
* use ontologies and controlled languages (thesaurus) that are standard in your field
---
class: left, middle
# And why I would share MY data?
.pull-left[
<img src="images/harpagon.jpg" width="600" />
]
.pull-right[
* This is a lot of work to collect them!
* These data are a competitive asset for my laboratory
* Loi Lemaire : the public pre-empts the data to prevent it from becoming the property of private actors (avoiding enclosures)
* Unlike her works, the data actually belong to the researcher's institution, they are not hers.
]
---
background-color: #bebebe
class: middle, center
# How to write a data management plan (DMP)
---
class: left, center
## DMP is compulsory for any research contract with National Research Agency (ANR)
## must be submitted 6 months after the signature of the contract
--
.red[
The DMP states what data will be collected and how they will be processed to be made FAIR compliant
]
---
class: left, middle
# where should you store your data?
.red[
** Not on the publisher's website**
]
* data repositories give you more place and freedom to describe your data
* publisher's policy may impede free reusability or sharing of your data (according to Open Science principles)
* publishers may have interest to make money with your data or use them to build products for sale (see for instance Elsevier and Scival)
---
class: left, middle
# Where should we store our data then?
* [re3data.org](https://re3data.org) (waiting for _recherche data gouv_ next summer)
* wiser to load them on a repository **before the publication of the related article** (once the paper is published, no possibility to make a link from the publisher website to the dataset on the repository)
it it's too late, you can still make a link from the open access repository (HAL) where you have deposited your AAM to the dataset.
* if your data are not ready to be deposited, you may nevertheless reserve a place on a data repository to get a permanent ID to the dataset by just loading a file (data description for instance)
* if you use your own data, provide a space before the bibliography entitled "data availability" for instance (and put here the DOI of the datasets)
* if you reuse date from another project, cite datasets in the bibliography
---
# consider the cost of data preservation
* The cost of data preservation should be balanced with the cost of production of a new dataset
* if your data are expensive to collect, you should archive them for longer, if the cost of preservation are higher than the cost of production, consider collecting new data
---
class: middle, left
# Store enough data to make reproducibility possible but no more :
* If your processes are well documented, you will be able to reproduce intermediary data from raw data (in this case, only raw data have to be stored in order to minimize the environmental footprint)
* computational data : virtualization (Docker) and versioning (github)
> every state of data that can be reproduced may not be stored. If we need intermediary data, and if every process is properly documented, it will only cost some computing time & resources to get them from the raw data and thus you will have spared the cost of storage for a large amount of refined data (Olivier Collin)
---
class: left, middle
# Why it is so important to describe your data
Write and make available at the root of your folder a readme file, with these informations:
- file organisation and localisation in the folder
- definition of the variables in the data and explanations about how the observations were conducted
- details on the experimentation settings
- details on the data processing (which operations, with what tools...)
---
class: middle, left
# description : ontologies and controlled languages.
## Ontologies :
- [genomic ontologies](http://geneontology.org/)
- [biological ontologies](https://www.ebi.ac.uk/ols/ontologies/uberon)
## thesaurus, controlled vocabularies :
- [MeSH](https://www.ncbi.nlm.nih.gov/mesh/?) (médecine)
- [INRAE thesaurus](https://consultation.vocabulaires-ouverts.inrae.fr/thesaurus-inrae/fr/)
---
class: left, center
## Documenting your data make them more reusable (and prevent bad interpretation)
Lack of context = no reusability
<img src="images/data_context.png" height="450" style="display: block; margin: auto;" />
---
class: left, middle
# data : from a privacy perspective
anonymisation # pseudonymisation
Tools: [Amnesia](https://amnesia.openaire.eu/download.html) et [Arx](https://arx.deidentifier.org/)
The more pseudonomysed personal data you have in a dataset, higher is the risk of a [reidentification](https://cpg.doc.ic.ac.uk/observatory/explore)
---
class: left, middle, inverse
# credits
This presentation was widely inspired by :
* Thierry Fournier, [introduction aux données de la recherche en sciences exactes](http://formadoct.doctorat-bretagneloire.fr/ld.php?content_id=33694689), 2022
* Cécile Arènes, [Rédiger un plan de gestion des données](https://zenodo.org/record/5559598), 2021
---
class: left, middle, inverse
# bibliography
<small>
Blanc, I. (2020, novembre 20). Données relatives à la concentration de SARS-CoV-2 dans les eaux usées—Une demande d’accès à l’information à Ministère de l’enseignement supérieur, de la recherche et de l’innovation. Consulté 14 avril 2022, à l’adresse Ma Dada website: https://madada.fr/demande/donnees_relatives_a_la_concentra
Brunori, G. (2020). Data Management Plan. https://doi.org/10.5281/zenodo.3664215
Colavizza, G., Hrynaszkiewicz, I., Staden, I., Whitaker, K., & McGillivray, B. (2020). The citation advantage of linking publications to research data. PLOS ONE, 15(4), e0230416. https://doi.org/10.1371/journal.pone.0230416
Munafò, M. R., Nosek, B. A., Bishop, D. V. M., Button, K. S., Chambers, C. D., Percie du Sert, N., … Ioannidis, J. P. A. (2017). A manifesto for reproducible science. Nature Human Behaviour, 1(1), 1‑9. https://doi.org/10.1038/s41562-016-0021
OCDE. (2007). Principes et lignes directrices de l’OCDE pour l’accès aux données de la recherche financée sur fonds publics. Consulté à l’adresse https://www.oecd.org/fr/science/inno/38500823.pdf
Sandve, G. K., Nekrutenko, A., Taylor, J., & Hovig, E. (2013). Ten Simple Rules for Reproducible Computational Research. PLOS Computational Biology, 9(10), e1003285. https://doi.org/10.1371/journal.pcbi.1003285
« The Lancet » annonce le retrait de son étude sur l’hydroxychloroquine. (2020, juin 4). Le Monde.fr. Consulté à l’adresse https://www.lemonde.fr/sciences/article/2020/06/04/hydroxychloroquine-trois-auteurs-de-l-etude-du-lancet-se-retractent_6041803_1650684.html
Vines, T. H., Albert, A. Y. K., Andrew, R. L., Débarre, F., Bock, D. G., Franklin, M. T., … Rennison, D. J. (2014). The Availability of Research Data Declines Rapidly with Article Age. Current Biology, 24(1), 94‑97. https://doi.org/10.1016/j.cub.2013.11.014
</small>
---
</textarea>
<style data-target="print-only">@media screen {.remark-slide-container{display:block;}.remark-slide-scaler{box-shadow:none;}}</style>
<script src="https://remarkjs.com/downloads/remark-latest.min.js"></script>
<script>var slideshow = remark.create({
"ratio": "16:9",
"highlightStyle": "github",
"highlightLines": true,
"countIncrementalSlides": false
});
if (window.HTMLWidgets) slideshow.on('afterShowSlide', function (slide) {
window.dispatchEvent(new Event('resize'));
});
(function(d) {
var s = d.createElement("style"), r = d.querySelector(".remark-slide-scaler");
if (!r) return;
s.type = "text/css"; s.innerHTML = "@page {size: " + r.style.width + " " + r.style.height +"; }";
d.head.appendChild(s);
})(document);
(function(d) {
var el = d.getElementsByClassName("remark-slides-area");
if (!el) return;
var slide, slides = slideshow.getSlides(), els = el[0].children;
for (var i = 1; i < slides.length; i++) {
slide = slides[i];
if (slide.properties.continued === "true" || slide.properties.count === "false") {
els[i - 1].className += ' has-continuation';
}
}
var s = d.createElement("style");
s.type = "text/css"; s.innerHTML = "@media print { .has-continuation { display: none; } }";
d.head.appendChild(s);
})(document);
// delete the temporary CSS (for displaying all slides initially) when the user
// starts to view slides
(function() {
var deleted = false;
slideshow.on('beforeShowSlide', function(slide) {
if (deleted) return;
var sheets = document.styleSheets, node;
for (var i = 0; i < sheets.length; i++) {
node = sheets[i].ownerNode;
if (node.dataset["target"] !== "print-only") continue;
node.parentNode.removeChild(node);
}
deleted = true;
});
})();
(function() {
"use strict"
// Replace <script> tags in slides area to make them executable
var scripts = document.querySelectorAll(
'.remark-slides-area .remark-slide-container script'
);
if (!scripts.length) return;
for (var i = 0; i < scripts.length; i++) {
var s = document.createElement('script');
var code = document.createTextNode(scripts[i].textContent);
s.appendChild(code);
var scriptAttrs = scripts[i].attributes;
for (var j = 0; j < scriptAttrs.length; j++) {
s.setAttribute(scriptAttrs[j].name, scriptAttrs[j].value);
}
scripts[i].parentElement.replaceChild(s, scripts[i]);
}
})();
(function() {
var links = document.getElementsByTagName('a');
for (var i = 0; i < links.length; i++) {
if (/^(https?:)?\/\//.test(links[i].getAttribute('href'))) {
links[i].target = '_blank';
}
}
})();
// adds .remark-code-has-line-highlighted class to <pre> parent elements
// of code chunks containing highlighted lines with class .remark-code-line-highlighted
(function(d) {
const hlines = d.querySelectorAll('.remark-code-line-highlighted');
const preParents = [];
const findPreParent = function(line, p = 0) {
if (p > 1) return null; // traverse up no further than grandparent
const el = line.parentElement;
return el.tagName === "PRE" ? el : findPreParent(el, ++p);
};
for (let line of hlines) {
let pre = findPreParent(line);
if (pre && !preParents.includes(pre)) preParents.push(pre);
}
preParents.forEach(p => p.classList.add("remark-code-has-line-highlighted"));
})(document);</script>
<script>
slideshow._releaseMath = function(el) {
var i, text, code, codes = el.getElementsByTagName('code');
for (i = 0; i < codes.length;) {
code = codes[i];
if (code.parentNode.tagName !== 'PRE' && code.childElementCount === 0) {
text = code.textContent;
if (/^\\\((.|\s)+\\\)$/.test(text) || /^\\\[(.|\s)+\\\]$/.test(text) ||
/^\$\$(.|\s)+\$\$$/.test(text) ||
/^\\begin\{([^}]+)\}(.|\s)+\\end\{[^}]+\}$/.test(text)) {
code.outerHTML = code.innerHTML; // remove <code></code>
continue;
}
}
i++;
}
};
slideshow._releaseMath(document);
</script>
<!-- dynamically load mathjax for compatibility with self-contained -->
<script>
(function () {
var script = document.createElement('script');
script.type = 'text/javascript';
script.src = 'https://mathjax.rstudio.com/latest/MathJax.js?config=TeX-MML-AM_CHTML';
if (location.protocol !== 'file:' && /^https?:/.test(script.src))
script.src = script.src.replace(/^https?:/, '');
document.getElementsByTagName('head')[0].appendChild(script);
})();
</script>
</body>
</html>