R: Community analysis I - Class Exercises
Jordan Casey
12/29/2022
1 Packages and data
Load packages and read in datasets for class exercises:
- Packages: tidyverse, ggplot2, fishualize, vegan
- Data: fishcoms_lizardisland.csv
lfish <- read.csv(file = "data/fishcoms_lizardisland.csv")2 Exercise 1
The “fishcoms_lizardisland.csv” dataset includes counts of reef fishes at fourteen sites around Lizard Island, Australia. The sites are shown in the map below.
The sites are marked with symbols, indicating different regimes
of wave action. The regimes of wave action include exposed, oblique,
sheltered, and lagoon, and they may impact reef fish community
composition. The data were collected by divers that ran transects at two
distinct depths to survey fishes at each site. For simplicity in this
exercise, we will focus on a single family, Pomacentridae, which are
commonly known as damselfishes.
2.1 Part I
- Filter the families to only retain members of the Pomacentridae.
- Calculate the sum of individuals (abundance) of each species at each site across exposure regimes. This should yield a dataset with 389 rows and 4 columns.
- Spread the data into a wide format, using the spread() function. There will be lots of “NAs.” Get rid of the NAs using the term “fill = 0” inside the spread() function.
# filter Pomacentridae and calculate the abundance of each species
lizard.sum <- lfish %>%
filter(Family == "Pomacentridae") %>%
group_by(Site, Exposure, species) %>%
summarize(abundance = sum(abundance))## `summarise()` has grouped output by 'Site', 'Exposure'. You can override using
## the `.groups` argument.lizard.sum## # A tibble: 389 × 4
## # Groups: Site, Exposure [14]
## Site Exposure species abundance
## <chr> <chr> <chr> <int>
## 1 Big Vickis Sheltered abu.whit 4
## 2 Big Vickis Sheltered acn.poly 23
## 3 Big Vickis Sheltered amb.cura 57
## 4 Big Vickis Sheltered amb.leuc 7
## 5 Big Vickis Sheltered amp.akin 3
## 6 Big Vickis Sheltered amp.clar 3
## 7 Big Vickis Sheltered chr.alis 12
## 8 Big Vickis Sheltered chr.tern 291
## 9 Big Vickis Sheltered chr.viri 79
## 10 Big Vickis Sheltered chy.roll 12
## # … with 379 more rows# spread into wide format and get rid of NAs
lizard.wide <- lizard.sum %>%
spread(key = species, value = abundance, fill = 0)
lizard.wide## # A tibble: 14 × 57
## # Groups: Site, Exposure [14]
## Site Expos…¹ abu.s…² abu.w…³ acn.p…⁴ amb.a…⁵ amb.c…⁶ amb.l…⁷ amp.a…⁸ amp.c…⁹
## <chr> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 Big … Shelte… 0 4 23 0 57 7 3 3
## 2 Blue… Lagoon 4 0 35 0 307 1 0 0
## 3 Bomm… Oblique 0 0 62 0 57 21 0 0
## 4 Coco… Exposed 0 45 181 0 64 13 0 0
## 5 Gran… Shelte… 1 0 9 0 1 0 0 1
## 6 LTMP1 Exposed 1 268 72 0 54 19 0 0
## 7 LTMP3 Exposed 0 43 182 1 72 18 4 0
## 8 Monk… Lagoon 0 0 1 0 82 33 0 0
## 9 Nort… Shelte… 0 0 55 0 23 2 2 0
## 10 Palf… Lagoon 0 0 77 0 192 0 0 0
## 11 Pidg… Exposed 0 0 118 0 47 14 0 0
## 12 Reso… Shelte… 0 0 29 0 1 0 2 4
## 13 Stev… Oblique 0 0 23 0 44 7 1 1
## 14 Wash… Oblique 0 0 39 0 18 0 0 0
## # … with 47 more variables: amp.mela <dbl>, amp.peri <dbl>, chr.alis <dbl>,
## # chr.lepi <dbl>, chr.marg <dbl>, chr.tern <dbl>, chr.viri <dbl>,
## # chr.webe <dbl>, chr.xant <dbl>, chy.cyan <dbl>, chy.flav <dbl>,
## # chy.rex <dbl>, chy.roll <dbl>, chy.talb <dbl>, das.arua <dbl>,
## # das.reti <dbl>, das.trim <dbl>, dis.mela <dbl>, dis.pers <dbl>,
## # dis.pros <dbl>, dis.pseu <dbl>, hgy.plag <dbl>, neg.mela <dbl>,
## # neg.nigr <dbl>, neo.azys <dbl>, neo.cyan <dbl>, pgy.dick <dbl>, …2.2 Part II
- Split your dataset into metadata (site and exposure information) and numeric data.
- Run a PCA and view the summary. Note the percent variation explained by PC1 and PC2.
# isolate metadata
lizard.meta <- lizard.wide[1:2]
lizard.meta## # A tibble: 14 × 2
## # Groups: Site, Exposure [14]
## Site Exposure
## <chr> <chr>
## 1 Big Vickis Sheltered
## 2 Blue Lagoon Lagoon
## 3 Bommie Bay Oblique
## 4 Coconut Exposed
## 5 Granite Sheltered
## 6 LTMP1 Exposed
## 7 LTMP3 Exposed
## 8 Monkey's Butt Lagoon
## 9 North Reef Sheltered
## 10 Palfrey Lagoon Lagoon
## 11 Pidgin Point Exposed
## 12 Resort Sheltered
## 13 Steve's Gully Oblique
## 14 Washing Machine Oblique# isolate numeric data
lizard.raw <- lizard.wide %>%
ungroup() %>%
select(-c(1:2))
lizard.raw## # A tibble: 14 × 55
## abu.sexs abu.whit acn.poly amb.aure amb.cura amb.leuc amp.a…¹ amp.c…² amp.m…³
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 0 4 23 0 57 7 3 3 0
## 2 4 0 35 0 307 1 0 0 0
## 3 0 0 62 0 57 21 0 0 4
## 4 0 45 181 0 64 13 0 0 0
## 5 1 0 9 0 1 0 0 1 0
## 6 1 268 72 0 54 19 0 0 0
## 7 0 43 182 1 72 18 4 0 0
## 8 0 0 1 0 82 33 0 0 0
## 9 0 0 55 0 23 2 2 0 0
## 10 0 0 77 0 192 0 0 0 0
## 11 0 0 118 0 47 14 0 0 0
## 12 0 0 29 0 1 0 2 4 0
## 13 0 0 23 0 44 7 1 1 0
## 14 0 0 39 0 18 0 0 0 0
## # … with 46 more variables: amp.peri <dbl>, chr.alis <dbl>, chr.lepi <dbl>,
## # chr.marg <dbl>, chr.tern <dbl>, chr.viri <dbl>, chr.webe <dbl>,
## # chr.xant <dbl>, chy.cyan <dbl>, chy.flav <dbl>, chy.rex <dbl>,
## # chy.roll <dbl>, chy.talb <dbl>, das.arua <dbl>, das.reti <dbl>,
## # das.trim <dbl>, dis.mela <dbl>, dis.pers <dbl>, dis.pros <dbl>,
## # dis.pseu <dbl>, hgy.plag <dbl>, neg.mela <dbl>, neg.nigr <dbl>,
## # neo.azys <dbl>, neo.cyan <dbl>, pgy.dick <dbl>, pgy.lacr <dbl>, …# run PCA and view summary
lizard.pca <- summary(rda(lizard.raw, scale = T))
lizard.pca##
## Call:
## rda(X = lizard.raw, scale = T)
##
## Partitioning of correlations:
## Inertia Proportion
## Total 55 1
## Unconstrained 55 1
##
## Eigenvalues, and their contribution to the correlations
##
## Importance of components:
## PC1 PC2 PC3 PC4 PC5 PC6 PC7
## Eigenvalue 11.9775 8.5486 6.1300 5.24090 4.55776 3.86037 3.44810
## Proportion Explained 0.2178 0.1554 0.1115 0.09529 0.08287 0.07019 0.06269
## Cumulative Proportion 0.2178 0.3732 0.4847 0.57995 0.66281 0.73300 0.79570
## PC8 PC9 PC10 PC11 PC12 PC13
## Eigenvalue 2.72451 2.68263 2.45025 1.40725 1.14087 0.83126
## Proportion Explained 0.04954 0.04878 0.04455 0.02559 0.02074 0.01511
## Cumulative Proportion 0.84523 0.89401 0.93856 0.96414 0.98489 1.00000
##
## Scaling 2 for species and site scores
## * Species are scaled proportional to eigenvalues
## * Sites are unscaled: weighted dispersion equal on all dimensions
## * General scaling constant of scores: 5.171023
##
##
## Species scores
##
## PC1 PC2 PC3 PC4 PC5 PC6
## abu.sexs -0.28419 -0.26752 -0.059225 -0.307390 0.337479 0.033509
## abu.whit 0.39862 -0.39324 -0.215994 -0.230512 -0.175736 -0.020574
## acn.poly 0.29575 -0.29428 0.156874 0.444956 0.061630 -0.126857
## amb.aure 0.17242 -0.05354 -0.041495 0.459242 0.072075 0.307651
## amb.cura -0.50387 -0.42986 -0.024245 0.011561 0.127535 0.012262
## amb.leuc 0.24375 -0.23786 0.136541 0.028162 -0.097926 0.373732
## amp.akin 0.08714 0.24320 -0.259432 0.352502 -0.110470 0.282233
## amp.clar -0.07053 0.49040 -0.282152 -0.076162 -0.170052 -0.106559
## amp.mela -0.01347 0.02095 0.515011 -0.073951 -0.271962 0.072117
## amp.peri 0.10372 -0.06009 0.176475 0.009064 0.273628 -0.125688
## chr.alis -0.32174 -0.13858 -0.158218 0.315366 -0.252461 -0.358979
## chr.lepi 0.12845 0.39671 -0.196386 -0.206075 0.178960 -0.244934
## chr.marg 0.14690 0.14568 0.076212 -0.045410 0.463959 -0.204090
## chr.tern -0.08762 0.15894 -0.162852 0.187699 -0.205306 0.396930
## chr.viri -0.31372 -0.17712 -0.220996 0.379202 -0.259790 -0.110179
## chr.webe 0.41034 -0.36908 -0.229689 -0.277514 -0.176346 -0.071793
## chr.xant 0.33785 -0.36208 -0.225573 -0.338710 -0.194370 -0.021957
## chy.cyan 0.33785 -0.36208 -0.225573 -0.338710 -0.194370 -0.021957
## chy.flav 0.17663 -0.03738 -0.048634 0.456347 0.078587 0.303659
## chy.rex 0.37729 -0.25346 0.158128 0.084928 -0.153207 -0.120968
## chy.roll -0.46780 -0.08951 -0.085420 -0.240278 0.193017 0.220226
## chy.talb 0.48315 -0.33436 -0.156873 -0.131655 -0.021849 -0.034676
## das.arua -0.14115 0.11983 0.468754 -0.088166 -0.286340 0.109579
## das.reti 0.26936 0.22029 0.025710 0.132111 0.294194 -0.016097
## das.trim 0.16418 0.21857 -0.081942 -0.037291 0.364127 -0.167839
## dis.mela -0.44746 -0.21703 0.380647 -0.195615 0.086775 0.073649
## dis.pers -0.33768 -0.01578 0.455202 -0.014191 -0.318797 0.057016
## dis.pros -0.61661 -0.24269 0.009592 -0.086497 0.096826 0.019691
## dis.pseu -0.32658 -0.15982 0.159176 0.183557 -0.317450 -0.246025
## hgy.plag -0.58366 -0.30691 -0.087950 -0.065793 0.151243 0.006203
## neg.mela -0.21683 0.21443 -0.394286 0.128747 -0.325971 -0.177912
## neg.nigr -0.07142 -0.48493 0.326830 0.127779 0.069717 -0.230511
## neo.azys -0.27253 -0.36310 -0.159683 0.391081 0.010686 0.083854
## neo.cyan -0.05513 0.37283 -0.150648 -0.154607 -0.098516 -0.161079
## pgy.dick 0.33785 -0.36208 -0.225573 -0.338710 -0.194370 -0.021957
## pgy.lacr 0.25495 -0.14912 0.542489 -0.072325 -0.117476 -0.010212
## pom.adel -0.51543 -0.03938 -0.186250 0.073550 -0.225056 0.102696
## pom.ambo -0.03098 0.42103 0.424132 -0.124514 -0.084066 0.002887
## pom.bank 0.25651 -0.17142 0.157411 0.231577 0.054708 -0.264610
## pom.brac 0.46792 -0.28082 -0.094161 -0.063335 -0.147129 0.360747
## pom.chry 0.04138 0.38423 0.228579 -0.104038 -0.229685 -0.165112
## pom.coel 0.31375 0.08344 0.058390 0.389115 0.239269 -0.023073
## pom.gram -0.51409 -0.33874 -0.051160 -0.128472 0.195373 -0.024673
## pom.lepi 0.46488 -0.32997 -0.005540 -0.127255 -0.086308 -0.193079
## pom.molu -0.47077 -0.27995 -0.160364 -0.113964 0.078918 0.181714
## pom.naga 0.05716 0.57896 -0.227224 -0.102932 -0.175195 -0.060764
## pom.pavo -0.01347 0.02095 0.515011 -0.073951 -0.271962 0.072117
## pom.phil 0.10372 -0.06009 0.176475 0.009064 0.273628 -0.125688
## pom.vaiu 0.20457 -0.14456 0.115550 0.202010 0.061314 -0.295638
## pom.ward -0.26503 -0.25922 -0.190424 0.278866 -0.270421 -0.359023
## pre.biac -0.49335 0.12984 -0.104803 -0.074152 -0.153793 -0.147121
## ste.apic 0.46126 -0.35932 0.062863 0.222298 -0.004796 0.030094
## ste.fasc -0.01405 0.12662 0.242576 -0.106959 -0.072540 0.031099
## ste.livi -0.09204 -0.01065 -0.065753 -0.074514 0.039723 0.367842
## ste.nigr -0.58723 -0.28640 0.081650 -0.039123 0.067851 -0.110306
##
##
## Site scores (weighted sums of species scores)
##
## PC1 PC2 PC3 PC4 PC5 PC6
## sit1 -0.93429 1.14681 -1.3318 0.48051 -1.3949 0.8130
## sit2 -2.84456 -1.76235 0.1020 -1.50702 2.1676 0.5541
## sit3 -0.09629 0.14975 3.6805 -0.52848 -1.9436 0.5154
## sit4 1.46195 -1.03306 0.8258 1.44365 0.4382 -2.1128
## sit5 0.68520 1.84393 -0.5211 -0.50571 2.5496 -1.0835
## sit6 2.41440 -2.58755 -1.6120 -2.42056 -1.3891 -0.1569
## sit7 1.23222 -0.38260 -0.2965 3.28194 0.5151 2.1986
## sit8 -0.65773 -0.07614 -0.4699 -0.53251 0.2839 2.6288
## sit9 0.51828 0.49694 -0.5933 -0.10762 0.2692 0.6831
## sit10 -2.54240 -1.35673 -0.8371 1.76191 -1.4142 -2.2513
## sit11 0.74122 -0.42939 1.2612 0.06478 1.9555 -0.8982
## sit12 -0.13795 2.79377 -1.0600 -0.95696 -0.8849 -1.2943
## sit13 0.26033 0.29175 -0.8810 0.29045 -0.6340 0.1817
## sit14 -0.10038 0.90488 1.7335 -0.76438 -0.5184 0.22222.3 Part III
- Extract the site scores, and use the bind_cols() function to merge your metdata with the site scores.
- Create convex hulls grouped by site exposure.
- Extract loadings from species scores.
- Plot your PCA results, using color and convex hulls to indicate the four exposure regimes. Include the species as loadings (it will be messy). Indicate the variation explained by the principal components axes.
# extract the site scores and merge with metadata
lizard.scores <- as.data.frame(lizard.pca$sites) %>%
bind_cols(lizard.meta)
lizard.scores## PC1 PC2 PC3 PC4 PC5 PC6
## sit1 -0.93428885 1.14680849 -1.3318275 0.48050919 -1.3948564 0.8130396
## sit2 -2.84455673 -1.76235287 0.1019847 -1.50702164 2.1675739 0.5541368
## sit3 -0.09629497 0.14974825 3.6804891 -0.52848372 -1.9435562 0.5153788
## sit4 1.46194921 -1.03305693 0.8257666 1.44365153 0.4381775 -2.1127524
## sit5 0.68520238 1.84392761 -0.5211412 -0.50570661 2.5496164 -1.0835098
## sit6 2.41440360 -2.58755188 -1.6120391 -2.42056355 -1.3890501 -0.1569157
## sit7 1.23221836 -0.38259696 -0.2965427 3.28193653 0.5150804 2.1986084
## sit8 -0.65773262 -0.07614061 -0.4699026 -0.53251175 0.2838743 2.6287586
## sit9 0.51827546 0.49693889 -0.5932970 -0.10761588 0.2692489 0.6830832
## sit10 -2.54239937 -1.35672736 -0.8371333 1.76191338 -1.4142322 -2.2512591
## sit11 0.74122380 -0.42939491 1.2611655 0.06477657 1.9554600 -0.8982195
## sit12 -0.13795320 2.79377332 -1.0600425 -0.95695704 -0.8849320 -1.2942955
## sit13 0.26033095 0.29174976 -0.8810285 0.29044831 -0.6340032 0.1816979
## sit14 -0.10037801 0.90487520 1.7335485 -0.76437533 -0.5184014 0.2222486
## Site Exposure
## sit1 Big Vickis Sheltered
## sit2 Blue Lagoon Lagoon
## sit3 Bommie Bay Oblique
## sit4 Coconut Exposed
## sit5 Granite Sheltered
## sit6 LTMP1 Exposed
## sit7 LTMP3 Exposed
## sit8 Monkey's Butt Lagoon
## sit9 North Reef Sheltered
## sit10 Palfrey Lagoon Lagoon
## sit11 Pidgin Point Exposed
## sit12 Resort Sheltered
## sit13 Steve's Gully Oblique
## sit14 Washing Machine Oblique# create convex hulls grouped by site exposure based on site scores
lizard.hulls <- lizard.scores %>%
group_by(Exposure) %>%
slice(chull(PC1, PC2))
lizard.hulls## # A tibble: 13 × 8
## # Groups: Exposure [4]
## PC1 PC2 PC3 PC4 PC5 PC6 Site Exposure
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <chr> <chr>
## 1 2.41 -2.59 -1.61 -2.42 -1.39 -0.157 LTMP1 Exposed
## 2 0.741 -0.429 1.26 0.0648 1.96 -0.898 Pidgin Point Exposed
## 3 1.23 -0.383 -0.297 3.28 0.515 2.20 LTMP3 Exposed
## 4 -2.84 -1.76 0.102 -1.51 2.17 0.554 Blue Lagoon Lagoon
## 5 -2.54 -1.36 -0.837 1.76 -1.41 -2.25 Palfrey Lagoon Lagoon
## 6 -0.658 -0.0761 -0.470 -0.533 0.284 2.63 Monkey's Butt Lagoon
## 7 0.260 0.292 -0.881 0.290 -0.634 0.182 Steve's Gully Oblique
## 8 -0.0963 0.150 3.68 -0.528 -1.94 0.515 Bommie Bay Oblique
## 9 -0.100 0.905 1.73 -0.764 -0.518 0.222 Washing Machine Oblique
## 10 0.518 0.497 -0.593 -0.108 0.269 0.683 North Reef Sheltered
## 11 -0.934 1.15 -1.33 0.481 -1.39 0.813 Big Vickis Sheltered
## 12 -0.138 2.79 -1.06 -0.957 -0.885 -1.29 Resort Sheltered
## 13 0.685 1.84 -0.521 -0.506 2.55 -1.08 Granite Sheltered# extract the loadings from species scores
lizard.loadings <- as.data.frame(lizard.pca$species) %>%
mutate(species = rownames(.))
lizard.loadings## PC1 PC2 PC3 PC4 PC5
## abu.sexs -0.28419327 -0.26752383 -0.059225088 -0.307390321 0.337479497
## abu.whit 0.39861947 -0.39324233 -0.215993975 -0.230511815 -0.175735748
## acn.poly 0.29574626 -0.29428240 0.156873657 0.444956221 0.061630243
## amb.aure 0.17242438 -0.05353681 -0.041495242 0.459241551 0.072075223
## amb.cura -0.50386600 -0.42985734 -0.024244512 0.011560848 0.127534660
## amb.leuc 0.24375159 -0.23785582 0.136541457 0.028161508 -0.097926340
## amp.akin 0.08713807 0.24320191 -0.259432036 0.352502086 -0.110470123
## amp.clar -0.07052897 0.49039979 -0.282152148 -0.076161959 -0.170052176
## amp.mela -0.01347456 0.02095428 0.515011036 -0.073950754 -0.271961920
## amp.peri 0.10371949 -0.06008525 0.176474954 0.009064189 0.273627621
## chr.alis -0.32173627 -0.13858143 -0.158217639 0.315366094 -0.252461012
## chr.lepi 0.12845052 0.39670838 -0.196386028 -0.206074790 0.178959675
## chr.marg 0.14690154 0.14567668 0.076211943 -0.045409501 0.463958587
## chr.tern -0.08761978 0.15893725 -0.162851723 0.187699259 -0.205306171
## chr.viri -0.31371521 -0.17711991 -0.220996121 0.379202299 -0.259789579
## chr.webe 0.41034064 -0.36907879 -0.229688964 -0.277514217 -0.176345653
## chr.xant 0.33784762 -0.36207627 -0.225572719 -0.338709585 -0.194369849
## chy.cyan 0.33784762 -0.36207627 -0.225572719 -0.338709585 -0.194369849
## chy.flav 0.17662542 -0.03737942 -0.048634293 0.456346780 0.078587053
## chy.rex 0.37728806 -0.25346430 0.158128024 0.084928439 -0.153207474
## chy.roll -0.46779936 -0.08951004 -0.085420015 -0.240278383 0.193017265
## chy.talb 0.48315353 -0.33436207 -0.156872787 -0.131655148 -0.021849360
## das.arua -0.14114944 0.11982924 0.468754200 -0.088166296 -0.286340297
## das.reti 0.26936274 0.22028882 0.025710178 0.132111229 0.294193936
## das.trim 0.16418492 0.21857114 -0.081941651 -0.037291421 0.364126691
## dis.mela -0.44746086 -0.21703150 0.380647325 -0.195614790 0.086774575
## dis.pers -0.33768073 -0.01578407 0.455202332 -0.014191389 -0.318797423
## dis.pros -0.61661247 -0.24269083 0.009591828 -0.086497132 0.096825901
## dis.pseu -0.32657673 -0.15982377 0.159175811 0.183557472 -0.317450009
## hgy.plag -0.58366260 -0.30690735 -0.087949915 -0.065792718 0.151242537
## neg.mela -0.21682695 0.21443247 -0.394286062 0.128746908 -0.325971020
## neg.nigr -0.07142109 -0.48492550 0.326829923 0.127779395 0.069716543
## neo.azys -0.27253308 -0.36310181 -0.159682667 0.391081209 0.010686048
## neo.cyan -0.05513116 0.37282574 -0.150647644 -0.154606936 -0.098516033
## pgy.dick 0.33784762 -0.36207627 -0.225572719 -0.338709585 -0.194369849
## pgy.lacr 0.25495205 -0.14912186 0.542488557 -0.072324509 -0.117475989
## pom.adel -0.51543119 -0.03938283 -0.186250081 0.073550387 -0.225056482
## pom.ambo -0.03098304 0.42102544 0.424131623 -0.124514067 -0.084065539
## pom.bank 0.25651170 -0.17142361 0.157410876 0.231577426 0.054707745
## pom.brac 0.46792295 -0.28082448 -0.094160546 -0.063335489 -0.147129336
## pom.chry 0.04137882 0.38423187 0.228579436 -0.104037870 -0.229684893
## pom.coel 0.31374852 0.08344050 0.058390166 0.389114846 0.239269183
## pom.gram -0.51408622 -0.33873889 -0.051160453 -0.128471569 0.195372569
## pom.lepi 0.46488159 -0.32997179 -0.005540028 -0.127255022 -0.086308343
## pom.molu -0.47076713 -0.27995181 -0.160363640 -0.113964393 0.078918117
## pom.naga 0.05715818 0.57895542 -0.227224109 -0.102931912 -0.175194614
## pom.pavo -0.01347456 0.02095428 0.515011036 -0.073950754 -0.271961920
## pom.phil 0.10371949 -0.06008525 0.176474954 0.009064189 0.273627621
## pom.vaiu 0.20457063 -0.14455571 0.115549561 0.202010234 0.061314200
## pom.ward -0.26503479 -0.25921597 -0.190424056 0.278866024 -0.270420530
## pre.biac -0.49334666 0.12984104 -0.104803086 -0.074152045 -0.153793321
## ste.apic 0.46125538 -0.35932241 0.062863410 0.222298460 -0.004795639
## ste.fasc -0.01404590 0.12661923 0.242575533 -0.106959081 -0.072539937
## ste.livi -0.09203656 -0.01065436 -0.065753489 -0.074514397 0.039722552
## ste.nigr -0.58723132 -0.28640372 0.081650061 -0.039123290 0.067850831
## PC6 species
## abu.sexs 0.033508875 abu.sexs
## abu.whit -0.020574399 abu.whit
## acn.poly -0.126856893 acn.poly
## amb.aure 0.307651394 amb.aure
## amb.cura 0.012262000 amb.cura
## amb.leuc 0.373731524 amb.leuc
## amp.akin 0.282233288 amp.akin
## amp.clar -0.106559193 amp.clar
## amp.mela 0.072116986 amp.mela
## amp.peri -0.125687908 amp.peri
## chr.alis -0.358978925 chr.alis
## chr.lepi -0.244934146 chr.lepi
## chr.marg -0.204089843 chr.marg
## chr.tern 0.396929932 chr.tern
## chr.viri -0.110179025 chr.viri
## chr.webe -0.071793253 chr.webe
## chr.xant -0.021957216 chr.xant
## chy.cyan -0.021957216 chy.cyan
## chy.flav 0.303658541 chy.flav
## chy.rex -0.120968236 chy.rex
## chy.roll 0.220225547 chy.roll
## chy.talb -0.034675977 chy.talb
## das.arua 0.109579294 das.arua
## das.reti -0.016097258 das.reti
## das.trim -0.167839287 das.trim
## dis.mela 0.073649307 dis.mela
## dis.pers 0.057016283 dis.pers
## dis.pros 0.019690868 dis.pros
## dis.pseu -0.246024920 dis.pseu
## hgy.plag 0.006202785 hgy.plag
## neg.mela -0.177911878 neg.mela
## neg.nigr -0.230511305 neg.nigr
## neo.azys 0.083854305 neo.azys
## neo.cyan -0.161078552 neo.cyan
## pgy.dick -0.021957216 pgy.dick
## pgy.lacr -0.010211576 pgy.lacr
## pom.adel 0.102696006 pom.adel
## pom.ambo 0.002886777 pom.ambo
## pom.bank -0.264610435 pom.bank
## pom.brac 0.360747040 pom.brac
## pom.chry -0.165111671 pom.chry
## pom.coel -0.023073342 pom.coel
## pom.gram -0.024672657 pom.gram
## pom.lepi -0.193078642 pom.lepi
## pom.molu 0.181714380 pom.molu
## pom.naga -0.060763511 pom.naga
## pom.pavo 0.072116986 pom.pavo
## pom.phil -0.125687908 pom.phil
## pom.vaiu -0.295637553 pom.vaiu
## pom.ward -0.359022962 pom.ward
## pre.biac -0.147120789 pre.biac
## ste.apic 0.030093941 ste.apic
## ste.fasc 0.031099259 ste.fasc
## ste.livi 0.367842329 ste.livi
## ste.nigr -0.110305800 ste.nigr# plot PCA results with color and convex hulls as exposure and species as loadings
lizard.plot <- ggplot() +
geom_point(data = lizard.scores, aes(x = PC1, y = PC2, color = Exposure), size = 2) +
geom_polygon(data = lizard.hulls, aes(x = PC1, y = PC2, fill = Exposure), alpha = 0.5) +
geom_segment(data = lizard.loadings, aes(x = 0, xend = PC1*3,
y = 0, yend = PC2*3), lwd = 0.1) +
geom_text(data = lizard.loadings, aes(x = PC1*3, y = PC2*3, label = species), size = 3) +
scale_fill_fish_d(option = "Stegastes_nigricans", end = 0.8) +
scale_color_fish_d(option = "Stegastes_nigricans", end = 0.8) +
theme_bw() +
xlab("PC1 (21.8%)") +
ylab("PC2 (15.5%)")## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.lizard.plot