/
session2.html
521 lines (421 loc) · 13.1 KB
/
session2.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
<!DOCTYPE html>
<html lang="" xml:lang="">
<head>
<title>Training in ade4 in R - Module I: Basic methods</title>
<meta charset="utf-8" />
<meta name="author" content="Stéphane Dray" />
<meta name="date" content="2023-12-06" />
<script src="libs/header-attrs/header-attrs.js"></script>
<link rel="stylesheet" href="custom.css" type="text/css" />
<link rel="stylesheet" href="xaringan-themer.css" type="text/css" />
</head>
<body>
<textarea id="source">
class: center, middle, inverse, title-slide
.title[
# Training in ade4 in R - Module I: Basic methods
]
.subtitle[
## Principal component analysis
]
.author[
### Stéphane Dray
]
.date[
### 2023-12-06
]
---
$$
%center text/code
\newcommand{\bcenter}{\begin{center}}
\newcommand{\ecenter}{\end{center}}
$$
$$
\newcommand{\bm}[1]{\boldsymbol{\mathbf{#1}}}
\newcommand{\tr}{\hspace{-0.05cm}^{\top}\hspace{-0.05cm}}% transpose d'une matrice
\newcommand{\mb}[1]{\mathbf{#1}}
\newcommand{\pt}{\mathsmaller \bullet}% petit point pour les indices
$$
$$
\newcommand{\triplet}[3]{
% pour ecrire un triplet dans le texte
$\left ( #1, #2, #3 \right )$
}
$$
$$
\newcommand{\sqnorm}[2]{
%norme au carré
\lVert #1 \rVert^2_{#2}
}
$$
$$
\newcommand{\norm}[2]{
%norme
\lVert #1 \rVert_{#2}
}
$$
---
# Data structure
.left-column[
<img src="img/onetable.png" width="156" style="display: block; margin: auto;" />
]
.right-column[
* One table with *p* variables measured on *n* individuals
* All variables are **quantitative**
* For instance
- sites `\(\times\)` environmental variables
- species `\(\times\)` traits
- individuals `\(\times\)` alleles
- populations `\(\times\)` alleles
]
---
# Objectives
* Identify what is the main information contained in the table
- Identify which variables are the most linked
- Identify the principal differences/similarities between individuals
---
# Data
We consider the `meaudret` data set
```r
library(ade4)
data(meaudret)
names(meaudret)
```
```
## [1] "env" "design" "spe" "spe.names"
```
```r
dim(meaudret$env)
```
```
## [1] 20 9
```
```r
names(meaudret$env)
```
```
## [1] "Temp" "Flow" "pH" "Cond" "Bdo5" "Oxyd" "Ammo" "Nitr" "Phos"
```
---
The data set contains an environmental table with 20 measurements of 9 environmental variables. The measurements have been made in 5 sites at each season along a small French stream (see `?meaudret`)
```r
head(meaudret$design)
```
```
## season site
## sp_1 spring S1
## sp_2 spring S2
## sp_3 spring S3
## sp_4 spring S4
## sp_5 spring S5
## su_1 summer S1
```
We want to know
* what are the main environmental gradients, i.e., which variables co-vary (if any)
* which samples have similar/different environmental conditions
---
# Principal component analysis
* `\(\mathbf{X}\)` contains centred or scaled variables
* `\(\mathbf{Q} = \mathbf{I}_p\)` is the identity matrix (diagonal matrix with 1s)
* `\(\mathbf{D} = \frac{1}{n}\mathbf{I}_n\)` is the diagonal matrix with `\(\frac{1}{n}\)`
.column-left[
<img src="img/onetable.png" width="156" style="display: block; margin: auto;" />
]
.column-center[
.center[
`dudi.pca`
<img src="img/arrow.png" width="131" style="display: block; margin: auto;" />
]
]
.column-right[
<img src="img/pca-map.png" width="353" style="display: block; margin: auto;" />
]
---
# Maximized criteria
* For individuals
$$ Q(\mathbf{a})=\sqnorm{\mathbf{XQa}}{\mb{D}} = \sqnorm{\mathbf{Xa}}{\frac{1}{n}\mb{I}_n} = var(\mathbf{Xa})= \lambda
$$
* For variables
- Centred data ( `\(x_{ij} - \bar{x}_j\)` )
`$$S(\mathbf{b})=\sqnorm{\mathbf{X}\tr\mathbf{Db}}{\mb{Q}} =\sqnorm{\frac{1}{n}\mathbf{X}\tr\mathbf{b}}{\mb{I}_p} = \sum_{j=1}^p{cov^2(\mathbf{x}_j, \mathbf{b})} = \lambda$$`
- Scaled data ( `\((x_{ij} - \bar{x}_j) / s_j\)` )
`$$S(\mathbf{b})=\sqnorm{\mathbf{X}\tr\mathbf{Db}}{\mb{Q}} =\sqnorm{\frac{1}{n}\mathbf{X}\tr\mathbf{b}}{\mb{I}_p} = \sum_{j=1}^p{cor^2(\mathbf{x}_j, \mathbf{b})} = \lambda$$`
---
# The `dudi.pca` function
## Arguments
```r
args(dudi.pca)
```
```
## function (df, row.w = rep(1, nrow(df))/nrow(df), col.w = rep(1,
## ncol(df)), center = TRUE, scale = TRUE, scannf = TRUE, nf = 2)
## NULL
```
* `df` is a `data.frame` with the data
* `row.w` and `col.w` are optional vectors of weights
* `center` and `scale` define the standardization of the data
* `scannf` and `nf` allow to set the number of dimensions to interpret
```r
pca.meau <- dudi.pca(meaudret$env, scannf = FALSE)
```
---
## Returned values
```r
names(pca.meau)
```
```
## [1] "tab" "cw" "lw" "eig" "rank" "nf" "c1" "li" "co" "l1" "call"
## [12] "cent" "norm"
```
It returns an object of class `dudi` containing:
- `$eig`: eigenvalues ( `\(\mb{\Lambda}\)` )
- `$cw`: column weights ( `\(\mb{Q}=\mb{I}_p\)` )
- `$lw`: row weights ( `\(\mb{D}=\frac{1}{n}\mb{I}_n\)` )
- `$tab`: transformed data table ( `\(\mb{X}\)` )
- `$c1`: principal axes or variable loadings ( `\(\mb{A}\)` )
- `$li`: row scores ( `\(\mb{L}=\mathbf{XA}\)` )
- `$l1`: principal components ( `\(\mb{B}\)` )
- `$co`: column scores ( `\(\mb{C}=\frac{1}{n}\mb{X}\tr\mb{B}\)` )
---
# Graphical representation and interpretation
As we have *two* analyses (individuals and variables spaces), two representations can be defined:
* **distance biplot** where `\(\mb{A}\)` and `\(\mb{L}=\mathbf{XA}\)` (`$c1`, `$li`) are superimposed.
* **correlation biplot** where `\(\mb{B}\)` and `\(\mb{C}=\frac{1}{n}\mb{X}\tr\mb{B}\)` (`$l1`, `$co`) are superimposed.
In the first interpretation, PCA finds coefficients for variables (`$c1`) to compute a linear combination (`$li`) that provides an ordination of individuals with the greatest dispersion (maximum variance).
In the second interpretation, PCA provides a linear combination (`$l1`) that maximise the correlations (`$co`) with all variables (or covariances for centred PCA). Hence, it is the best summary of the variables.
---
## The `biplot` function
```r
library(adegraphics)
```
.pull-left[
```r
biplot(pca.meau)
```
<img src="fig/unnamed-chunk-11-1.svg" style="display: block; margin: auto;" />
]
.pull-right[
```r
biplot(pca.meau, permute = TRUE)
```
<img src="fig/unnamed-chunk-12-1.svg" style="display: block; margin: auto;" />
]
---
## Separate representations
.pull-left[
```r
s.label(pca.meau$li)
```
<img src="fig/unnamed-chunk-13-1.svg" style="display: block; margin: auto;" />
]
.pull-right[
```r
s.arrow(pca.meau$co)
```
<img src="fig/unnamed-chunk-14-1.svg" style="display: block; margin: auto;" />
]
---
## Separate representations
.pull-left[
```r
s.label(pca.meau$li)
```
<img src="fig/unnamed-chunk-15-1.svg" style="display: block; margin: auto;" />
]
.pull-right[
```r
s.corcircle(pca.meau$co)
```
<img src="fig/unnamed-chunk-16-1.svg" style="display: block; margin: auto;" />
]
---
## Separate representations
.pull-left[
```r
s.class(pca.meau$li, meaudret$design[, 1], col = TRUE)
```
<img src="fig/unnamed-chunk-17-1.svg" style="display: block; margin: auto;" />
]
.pull-right[
```r
s.arrow(pca.meau$co)
```
<img src="fig/unnamed-chunk-18-1.svg" style="display: block; margin: auto;" />
]
---
# To scale or not to scale
Scaling should be performed when we do not want that differences in variances affect the results
```r
pca.meau.c <- dudi.pca(meaudret$env, scannf = FALSE,
scale = FALSE)
```
.pull-left[
<img src="fig/unnamed-chunk-20-1.svg" width="70%" style="display: block; margin: auto;" />
]
.pull-right[
<img src="fig/unnamed-chunk-21-1.svg" width="70%" style="display: block; margin: auto;" />
]
In our case, we must scale the data as differences in variances are mainly due to differences in units
---
# Inertia statistics
```r
summary(pca.meau)
```
```
## Class: pca dudi
## Call: dudi.pca(df = meaudret$env, scannf = FALSE)
##
## Total inertia: 9
##
## Eigenvalues:
## Ax1 Ax2 Ax3 Ax4 Ax5
## 5.1747 1.3204 1.0934 0.7321 0.4902
##
## Projected inertia (%):
## Ax1 Ax2 Ax3 Ax4 Ax5
## 57.497 14.671 12.149 8.135 5.447
##
## Cumulative projected inertia (%):
## Ax1 Ax1:2 Ax1:3 Ax1:4 Ax1:5
## 57.50 72.17 84.32 92.45 97.90
##
## (Only 5 dimensions (out of 9) are shown)
```
---
# PCA in practice
.center[
[Go to practical 2](../../practical/session2/session2.html)
]
</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({
"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;
});
})();
// add `data-at-shortcutkeys` attribute to <body> to resolve conflicts with JAWS
// screen reader (see PR #262)
(function(d) {
let res = {};
d.querySelectorAll('.remark-help-content table tr').forEach(tr => {
const t = tr.querySelector('td:nth-child(2)').innerText;
tr.querySelectorAll('td:first-child .key').forEach(key => {
const k = key.innerText;
if (/^[a-z]$/.test(k)) res[k] = t; // must be a single letter (key)
});
});
d.body.setAttribute('data-at-shortcutkeys', JSON.stringify(res));
})(document);
(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>