-
Notifications
You must be signed in to change notification settings - Fork 3
/
Copy path.Rhistory
512 lines (512 loc) · 28 KB
/
.Rhistory
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
#####calling the function for all four names- matching each 'wrongname' against all of the 'rightnames'
for(i in 1:dim(nametest)[1])
{
nametest$genusmatches[i]=bestmatch(nametest$WrongGen[i],nametest$GoodGen)
nametest$sppmatches[i]=bestmatch(nametest$WrongSpp[i],nametest$GoodSpp)
}
nametest
library(tcltk2)
library(tcltk)
library(tcltk)
library(tcltk2)
library("metagear", lib.loc="C:/Program Files/R/R-3.4.0/library")
library(metagear)
biocLite('EBImage')
source("https://bioconductor.org/biocLite.R")
biocLite('EBImage')
library(metagear)
library(EBImage)
library(metagear)
?plot_PRISMA
?abstract_screener
library(metagear)
?PDFs_collect
?abstract_screener
?effort_initialize
data(example_references_metagear)
head(example_references_metagear)
write.csv(example_references_metagear,'~/gleon/Geisha/metagear_ex.csv')
summary(lm)
.011/0.017
5/8
176+120+18
314/40
load("C:/Users/vpatil/Documents/gleon/Geisha/phyto_package/github/MFG_CSR.rda")
View(mfg.csr)
View(mfg.csr)
load("C:/Users/vpatil/Documents/gleon/Geisha/phyto_package/github/sppMFG.rda")
print(ipm.lsgo,digits=3)
44693*.02
44693*.03
44693*.04
#------
# CSR validation
#------
library(ggplot2)
library(scales)
library(vegan)
rm(list = ls())
setwd('~/gleon/Geisha/phyto_package/GEISHA_phytoplankton_github_shared/')
dat.m = read.csv("madgwick_csr_data.csv",stringsAsFactors = F)
source('species_mfg_convert02042018.r')
source('mfg_csr.r')
dat.m$SV=dat.m$surface.area/dat.m$volume
dat.m$MSV=dat.m$mld*dat.m$SV
dat.m<-genus.species.extract(dat.m,'Species')
dat.m<-phyto.convert.df(dat.m)
dat.m$MFG[dat.m$Species=='Osscilatoria']='5a-FilaCyano'
dat.m$MFG[dat.m$Species=='Elaktothrix']='11b-GelaChlor'
dat.m$MFG.number=sapply(dat.m$MFG,function(x) strsplit(gsub('-','_',x),split='_')[[1]][1])
dat.m=mfg.csr.convert.df(dat.m,'MFG.number')
dat.m$CSR[dat.m$Species=='Small non-flagellates'|dat.m$Species=='Small flagellates']='C'
dat.m
s.group=dat.m[dat.m$CSR=='S',]
#final plot
csr.plot=ggplot(dat.m, aes(x = MSV, y = SV, color = CSR)) +
geom_point(size=3,shape=19,position=position_jitter(width=0.1,height=0.1))+
theme(text=element_text(face="bold",size=20))+
scale_x_log10(breaks = trans_breaks("log10", function(x) 10^x),
labels = trans_format("log10", math_format(10^.x)),limits=c(2,1000)) +
scale_y_log10(breaks = trans_breaks("log10", function(x) 10^x),
labels = trans_format("log10", math_format(10^.x)),limits=c(0.01,10))+
xlab("msv-1")+ylab("sv-1")+
annotate("text",x=10,y=10,label="C",fontface=2,size=10)+
annotate("text",x=1000,y=10,label="R",fontface=2,size=10)+
annotate("text",x=50,y=0.02,label="S",fontface=2,size=10)
#geom_text(aes(label = sh.name), hjust = 0, vjust = 0)
csr.plot
setwd('~/gleon/Geisha/phyto_package/mee ms/')
mfg.mugg<-read.csv('MFG_in_out_mugg.csv')
head(mfg.mugg)
confusion=table(mfg.mugg$MFG,mfg.mugg$MFG_new)
write.csv(confusion,'mugg_mfg_confusion.csv')
table(mfg.mugg$MFG==mfg.mugg$MFG_new)/dim(mfg.mugg)[1]
table(mfg.mugg$MFG==mfg.mugg$MFG_new)
mfg.mugg<-read.csv('MFG_in_out_mugg.csv',stringsAsFactors = T)
confusion=table(mfg.mugg$MFG,mfg.mugg$MFG_new)
write.csv(confusion,'mugg_mfg_confusion.csv')
table(mfg.mugg$MFG==mfg.mugg$MFG_new)/dim(mfg.mugg)[1]
mfg.mugg<-read.csv('MFG_in_out_mugg.csv',stringsAsFactors = F)
confusion=table(mfg.mugg$MFG,mfg.mugg$MFG_new)
write.csv(confusion,'mugg_mfg_confusion.csv')
table(mfg.mugg$MFG==mfg.mugg$MFG_new)/dim(mfg.mugg)[1]
table(is.na(mfg.mugg$MFG))
3/163
table(mfg.mugg$MFG==mfg.mugg$MFG_new)
table(mfg.mugg$Cause.of.Difference)
setwd('~/gleon/Geisha/phyto_package/mee ms/')
mfg.mugg<-read.csv('MFG_in_out_mugg.csv',stringsAsFactors = F)
confusion=table(mfg.mugg$MFG,mfg.mugg$MFG_new)
write.csv(confusion,'mugg_mfg_confusion.csv')
table(mfg.mugg$MFG==mfg.mugg$MFG_new)/dim(mfg.mugg)[1]
table(mfg.mugg$MFG==mfg.mugg$MFG_new)
table(mfg.mugg$Cause.of.Difference)
table(is.na(mfg.mugg$MFG))
setwd('~/gleon/Geisha/phyto_package/mee ms/')
mfg.mugg<-read.csv('MFG_in_out_mugg.csv',stringsAsFactors = F)
confusion=table(mfg.mugg$MFG,mfg.mugg$MFG_new)
write.csv(confusion,'mugg_mfg_confusion.csv')
table(mfg.mugg$MFG==mfg.mugg$MFG_new)/dim(mfg.mugg)[1]
table(mfg.mugg$MFG==mfg.mugg$MFG_new)
table(mfg.mugg$Cause.of.Difference)
table(is.na(mfg.mugg$MFG))
setwd('~/gleon/Geisha/phyto_package/mee ms/')
mfg.mugg<-read.csv('MFG_in_out_mugg.csv',stringsAsFactors = F)
confusion=table(mfg.mugg$MFG,mfg.mugg$MFG_new)
write.csv(confusion,'mugg_mfg_confusion.csv')
table(mfg.mugg$MFG==mfg.mugg$MFG_new)/dim(mfg.mugg)[1]
table(mfg.mugg$MFG==mfg.mugg$MFG_new)
table(mfg.mugg$Cause.of.Difference)
16/19
2/19
1/19
16/21
###don't save- only written to create lake-specific temporary missing species lists for expert mfg classification filling
##there is something wrong with what you are doing- need to re-run the validation of nico's model versus trait-based classification and make sure you get the same correspondence that torsten described.
setwd('~/../Desktop/random files for mfg debug/')
source('species_mfg_convert02042018.r')
source('mfg_csr.r')
species.mfg.library=species.mfg.library[species.mfg.library$source=='Nico',] ###need to make sure you match this to the true original library, then you can send nico the library extension based on the best available size criteria.
source('~/../Desktop/random files for mfg debug/traits_to_mfg02042018.R')
setwd('~/gleon/Geisha/datasets/rawphytodata/phytodata_qaqc/')
mfg.traits.dbase<-read.csv('../../../phyto_package/github/orlane_fortraitstomfg.csv',stringsAsFactors = F)
species.mfg.library=species.mfg.library[species.mfg.library$source=='Nico',] ###need to make sure you match this to the true original library, then you can send nico the library extension based on the best available size criteria.
source('~/../Desktop/random files for mfg debug/traits_to_mfg02042018.R')
setwd('~/gleon/Geisha/datasets/rawphytodata/phytodata_qaqc/')
mfg.traits.dbase<-read.csv('../../../phyto_package/github/orlane_fortraitstomfg.csv',stringsAsFactors = F)
frederic.new=read.csv('~/../Desktop/random files for mfg debug/Appendix-1-Phytoplankton metrics database-revised.csv',stringsAsFactors = F)
setwd('~/gleon/Geisha/phyto_package/GEISHA_phytoplankton_github_shared/')
source('species_mfg_convert02042018.r')
source('mfg_csr.r')
species.mfg.library=species.mfg.library[species.mfg.library$source=='Nico',] ###need to make sure you match this to the true original library, then you can send nico the library extension based on the best available size criteria.
source('traits_to_mfg02042018.R')
mfg.traits.dbase<-read.csv('../github/orlane_fortraitstomfg.csv',stringsAsFactors = F)
frederic.new=read.csv('../mee ms/Appendix-1-Phytoplankton metrics database-revised.csv',stringsAsFactors = F)
names(frederic.new)[1]='Taxa.name'
mfg.traits.dbase=subset(mfg.traits.dbase,select=c('Taxa.name','aerotopes','Gelatinous','Centric'))
names(frederic.new)[1]='Taxa.name'
mfg.traits.dbase=subset(mfg.traits.dbase,select=c('Taxa.name','aerotopes','Gelatinous','Centric'))
mfg.traits.dbase$phyto_name=mfg.traits.dbase$Taxa.name
frederic.new<-merge(frederic.new,mfg.traits.dbase,all.x=T)
frederic.new<-subset(frederic.new,select=c("Class","Order","Mobility.apparatus","Heterotrophic","Mixotrophic","Autotrophic","Nano.microphytoplankton","Filament","Colonial","Geometrical.shape.of.the.cell","Gelatinous","aerotopes","Centric","phyto_name","Cell.surface.area.µm²","Cell.biovolume.µm3","Number.of.cells.per.colony","Cumulated.surface.area.of.cells.in.a.colony.µm²","Colony.length..takes.into.account.mucilage..µm","Colony.width..takes.into.account.mucilage..µm","Colony.thickness..takes.into.account.mucilage..µm","Colony.surface.area.µm²..with.mucilage.","Colony.biovolume.µm3..without.mucilage.","Maximal.length.of.the.algal.object"))
names(frederic.new)
f.newcols=c("Class","Order","Mobility.apparatus","Heterotrophic","Mixotrophic","Autotrophic","Nano.microphytoplankton","Filament","Colonial","Geometrical.shape.of.the.cell","Gelatinous","aerotopes","Centric","phyto_name","Cell.surface.area.µm²","Cell.biovolume.µm3","Number.of.cells.per.colony","Cumulated.surface.area.of.cells.in.a.colony.µm²","Colony.length..takes.into.account.mucilage..µm","Colony.width..takes.into.account.mucilage..µm","Colony.thickness..takes.into.account.mucilage..µm","Colony.surface.area.µm²..with.mucilage.","Colony.biovolume.µm3..without.mucilage.","Maximal.length.of.the.algal.object")
f.newcols[f.newcols %in% names(frederic.new)==F]
names(frederic.new)
f.newcols[f.newcols %in% names(frederic.new)==F]
names(frederic.new)
f.newcols[f.newcols %in% names(frederic.new)==F]
grep(frederic.new,'Cell.',values=T)
grep(frederic.new,'Cell.',value=T)
?grep
grep(frederic.new,'Cell.',value=T,fixed=T)
grep('Cell.',frederic.new,fixed=T,value=T)
grep('Cell.',names(frederic.new),value=T)
f.newcols=c("Class","Order","Mobility.apparatus","Heterotrophic","Mixotrophic","Autotrophic","Nano.microphytoplankton","Filament","Colonial","Geometrical.shape.of.the.cell","Gelatinous","aerotopes","Centric","phyto_name","Cell.surface.area.µm²","Cell.biovolume.µm3","Number.of.cells.per.colony","Cumulated.surface.area.of.cells.in.a.colony.µm²","Colony.length..takes.into.account.mucilage..µm","Colony.width..takes.into.account.mucilage..µm","Colony.thickness..takes.into.account.mucilage..µm","Colony.surface.area.µm²..with.mucilage.","Colony.biovolume.µm3..without.mucilage.","Maximal.length.of.the.algal.object")
f.newcols=c("Class","Order","Mobility.apparatus","Heterotrophic","Mixotrophic","Autotrophic","Nano.microphytoplankton","Filament","Colonial","Geometrical.shape.of.the.cell","Gelatinous","aerotopes","Centric","phyto_name","Cell.surface.area.µm²","Cell.biovolume.µm3","Number.of.cells.per.colony","Cumulated.surface.area.of.cells.in.a.colony.µm²","Colony.length..takes.into.account.mucilage..µm","Colony.width..takes.into.account.mucilage..µm","Colony.thickness..takes.into.account.mucilage..µm","Colony.surface.area.µm²..with.mucilage.","Colony.biovolume.µm3..without.mucilage.","Maximal.length.of.the.algal.object")
f.newcols[f.newcols %in% names(frederic.new)==F]
frederic.new<-subset(frederic.new,select=f.newcols)
frederic.new=frederic.new[!is.na(frederic.new$phyto_name),]
frederic.new$size=ifelse(frederic.new$Colonial==1 | frederic.new$Maximal.length.of.the.algal.object>63,'large','small')
frederic.new
dim(frederic.new)
mfg.traits.dbase=frederic.new
mfg.traits.dbase$trait.MFG=''
for(i in 1:dim(mfg.traits.dbase)[1])
{
mfg.traits.dbase$trait.MFG[i]=traits.to.mfg(flagella=mfg.traits.dbase$Mobility.apparatus[i],size=mfg.traits.dbase$size[i],colonial=mfg.traits.dbase$Colonial[i],filament=mfg.traits.dbase$Filament[i],centric=mfg.traits.dbase$Centric[i],gelatinous = mfg.traits.dbase$Gelatinous[i],aerotypes = mfg.traits.dbase$aerotopes[i],class=mfg.traits.dbase$Class[i],order=mfg.traits.dbase$Order[i])
print(i)
}
names(mfg.traits.dbase)
mfg.traits.dbase=genus.species.extract(mfg.traits.dbase,'phyto_name')
mfg.traits.dbase$species=gsub('cf.','',mfg.traits.dbase$species)
mfg.traits.dbase<-mfg.traits.dbase[!duplicated(mfg.traits.dbase),]
mfg.traits.dbase<-phyto.convert.df(mfg.traits.dbase)
mfg.traits.dbase$MFG=factor(mfg.traits.dbase$MFG,levels=unique(as.character(mfg.traits.dbase$MFG)))
mfg.traits.dbase$trait.MFG=factor(mfg.traits.dbase$trait.MFG,levels=unique(as.character(mfg.traits.dbase$MFG)))
##validating trait and expert classifications
confusion.mat=table(mfg.traits.dbase$MFG,mfg.traits.dbase$trait.MFG)
write.csv(confusion.mat,'~/../Desktop/random files for mfg debug/mfg_confusion_newlarge.csv')
confusion.mat
mfg.traits.dbase$MFG
View(mfg.traits.dbase)
##size cutoff modeling
mfg.traits.dbase$large.MFG=NA
mfg.traits.dbase$large.MFG[grep(c('1|6|8'),mfg.traits.dbase$MFG)]='L'
mfg.traits.dbase$large.MFG[grep(c('2|7|9'),mfg.traits.dbase$MFG)]='S'
size.vars=names(mfg.traits.dbase)[c(7,15,16,17,9,18,22,23,24)]
size.dat=mfg.traits.dbase[,size.vars]
sapply(size.dat,function(x) sum(is.na(x)))
size.dat$Colony.surface.area.µm²..with.mucilage.[is.na(size.dat$Colony.surface.area.µm²..with.mucilage.)]=size.dat$Cell.surface.area.µm²[is.na(size.dat$Colony.surface.area.µm²..with.mucilage.)]
size.dat$Colony.biovolume.µm3..without.mucilage.[is.na(size.dat$Colony.biovolume.µm3..without.mucilage.)]=size.dat$Cell.biovolume.µm3[is.na(size.dat$Colony.biovolume.µm3..without.mucilage.)]
row.names(size.dat)=mfg.traits.dbase$phyto_name
size.dat$large.MFG=mfg.traits.dbase$large.MFG
size.dat$large.MFG=mfg.traits.dbase$large.MFG
size.dat$large.MFG=mfg.traits.dbase$large.MFG
size.dat$large.MFG=ifelse(size.dat$large.MFG=='L',1,0)
size.dat$Nano.microphytoplankton=ifelse(size.dat$Nano.microphytoplankton=='Nanophytoplankton',0,1)
size.dat$Colonial[size.dat$Number.of.cells.per.colony>=2]=1
size.vars=names(mfg.traits.dbase)[c(7,15,16,17,9,18,22,23,24)]
size.dat=mfg.traits.dbase[,size.vars]
sapply(size.dat,function(x) sum(is.na(x)))
size.dat$Colony.surface.area.µm²..with.mucilage.[is.na(size.dat$Colony.surface.area.µm²..with.mucilage.)]=size.dat$Cell.surface.area.µm²[is.na(size.dat$Colony.surface.area.µm²..with.mucilage.)]
size.dat$Colony.biovolume.µm3..without.mucilage.[is.na(size.dat$Colony.biovolume.µm3..without.mucilage.)]=size.dat$Cell.biovolume.µm3[is.na(size.dat$Colony.biovolume.µm3..without.mucilage.)]
size.dat$large.MFG=mfg.traits.dbase$large.MFG
size.dat$large.MFG=ifelse(size.dat$large.MFG=='L',1,0)
size.dat$Nano.microphytoplankton=ifelse(size.dat$Nano.microphytoplankton=='Nanophytoplankton',0,1)
size.dat$Colonial[size.dat$Number.of.cells.per.colony>=2]=1
size.dat$Cell.surface.area.µm²=as.numeric(size.dat$Cell.surface.area.µm²)
size.dat$Cumulated.surface.area.of.cells.in.a.colony.µm²=as.numeric(size.dat$Cumulated.surface.area.of.cells.in.a.colony.µm²)
size.dat$Colony.surface.area.µm²..with.mucilage.=as.numeric(size.dat$Colony.surface.area.µm²..with.mucilage.)
library(randomForest)
size.dat<-na.omit(size.dat)
size.rf<-randomForest(large.MFG~.,data=size.dat,family='binomial')
mfg.traits.dbase$large.MFG=NA
mfg.traits.dbase$large.MFG[grep(c('1|6|8'),mfg.traits.dbase$MFG)]='L'
mfg.traits.dbase$large.MFG[grep(c('2|7|9'),mfg.traits.dbase$MFG)]='S'
# mfg.traits.dbase$diatom=0
# mfg.traits.dbase$diatom[grep(c('6|7'),mfg.traits.dbase$MFG)]=1
# mfg.traits.dbase$cyano=0
# mfg.traits.dbase$cyano[grep(c('4|5'),mfg.traits.dbase$MFG)]=1
# mfg.traits.dbase$flagellate=0
# mfg.traits.dbase$flagellate[grep(c('1|2|3'),mfg.traits.dbase$flagellate)]=1
size.vars=names(mfg.traits.dbase)[c(7,15,16,17,9,18,22,23,24)]
size.dat=mfg.traits.dbase[,size.vars]
sapply(size.dat,function(x) sum(is.na(x)))
size.dat$Colony.surface.area.µm²..with.mucilage.[is.na(size.dat$Colony.surface.area.µm²..with.mucilage.)]=size.dat$Cell.surface.area.µm²[is.na(size.dat$Colony.surface.area.µm²..with.mucilage.)]
size.dat$Colony.biovolume.µm3..without.mucilage.[is.na(size.dat$Colony.biovolume.µm3..without.mucilage.)]=size.dat$Cell.biovolume.µm3[is.na(size.dat$Colony.biovolume.µm3..without.mucilage.)]
size.dat$large.MFG=mfg.traits.dbase$large.MFG
size.dat$large.MFG=ifelse(size.dat$large.MFG=='L',1,0)
size.dat$Nano.microphytoplankton=ifelse(size.dat$Nano.microphytoplankton=='Nanophytoplankton',0,1)
size.dat$Colonial[size.dat$Number.of.cells.per.colony>=2]=1
size.dat$Cell.surface.area.µm²=as.numeric(size.dat$Cell.surface.area.µm²)
size.dat$Cumulated.surface.area.of.cells.in.a.colony.µm²=as.numeric(size.dat$Cumulated.surface.area.of.cells.in.a.colony.µm²)
size.dat$Colony.surface.area.µm²..with.mucilage.=as.numeric(size.dat$Colony.surface.area.µm²..with.mucilage.)
dim(size.dat)
head(size.dat)
table(size.dat$large.MFG)
mfg.traits.dbase$MFG
names(mfg.traits.dbase)
mfg.traits.dbase$genus
mfg.traits.dbase$species
mfg.traits.dbase<-phyto.convert.df(mfg.traits.dbase)
mfg.traits.dbase$MFG
phyto.convert.df
source('species_mfg_convert02042018.r')
head(mfg.traits.dbase)
tail(mfg.traits.dbase)
genus='Woronichinia'
species='naegeliana'
species.mfg.library
grep('Woronichinia',species.mfg.library$genus)
dim(species.mfg.library)
mfg.traits.dbase$phyto_name
grep('Scenedesmus',species.mfg.library$genus)
genus='Scenedesmus'
species='smithii'
species.phyto.convert(genus,species)
phyto.convert.df
phyto.df=mfg.traits.dbase
mfgs<-vector(length=dim(phyto.df)[1])
for(i in 1:dim(phyto.df)[1])
{
mfgs[i]=species.phyto.convert(phyto.df$genus[i],phyto.df$species[i])
}
mfgs
phyto.convert.df
mfgs<-vector(length=dim(phyto.df)[1])
for(i in 1:dim(phyto.df)[1])
{
mfgs[i]=species.phyto.convert(phyto.df$genus[i],phyto.df$species[i])
}
phyto.df$MFG=mfgs
#phyto.df$MFG.numb
phyto.df$MFG
mfg.traits.dbase$MFG
mfg.traits.dbase<-phyto.convert.df(mfg.traits.dbase)
mfg.traits.dbase$MFG
mfg.traits.dbase$trait.MFG
mfg.traits.dbase$trait.MFG=''
for(i in 1:dim(mfg.traits.dbase)[1])
{
mfg.traits.dbase$trait.MFG[i]=traits.to.mfg(flagella=mfg.traits.dbase$Mobility.apparatus[i],size=mfg.traits.dbase$size[i],colonial=mfg.traits.dbase$Colonial[i],filament=mfg.traits.dbase$Filament[i],centric=mfg.traits.dbase$Centric[i],gelatinous = mfg.traits.dbase$Gelatinous[i],aerotypes = mfg.traits.dbase$aerotopes[i],class=mfg.traits.dbase$Class[i],order=mfg.traits.dbase$Order[i])
print(i)
}
mfg.traits.dbase$trait.MFG
mfg.traits.dbase$MFG=factor(mfg.traits.dbase$MFG,levels=unique(as.character(mfg.traits.dbase$MFG)))
mfg.traits.dbase$trait.MFG=factor(mfg.traits.dbase$trait.MFG,levels=unique(as.character(mfg.traits.dbase$MFG)))
confusion.mat=table(mfg.traits.dbase$MFG,mfg.traits.dbase$trait.MFG)
confusion.mat
table(mfg.traits.dbase$MFG==mfg.traits.dbase$trait.MFG)
632/(632+190)
size.vars=names(mfg.traits.dbase)[c(7,15,16,17,9,18,22,23,24)]
size.dat=mfg.traits.dbase[,size.vars]
sapply(size.dat,function(x) sum(is.na(x)))
size.dat$Colony.surface.area.µm²..with.mucilage.[is.na(size.dat$Colony.surface.area.µm²..with.mucilage.)]=size.dat$Cell.surface.area.µm²[is.na(size.dat$Colony.surface.area.µm²..with.mucilage.)]
size.dat$Colony.biovolume.µm3..without.mucilage.[is.na(size.dat$Colony.biovolume.µm3..without.mucilage.)]=size.dat$Cell.biovolume.µm3[is.na(size.dat$Colony.biovolume.µm3..without.mucilage.)]
size.dat$large.MFG=mfg.traits.dbase$large.MFG
size.dat$large.MFG=ifelse(size.dat$large.MFG=='L',1,0)
size.dat$Nano.microphytoplankton=ifelse(size.dat$Nano.microphytoplankton=='Nanophytoplankton',0,1)
size.dat$Colonial[size.dat$Number.of.cells.per.colony>=2]=1
size.dat$Cell.surface.area.µm²=as.numeric(size.dat$Cell.surface.area.µm²)
size.dat$Cumulated.surface.area.of.cells.in.a.colony.µm²=as.numeric(size.dat$Cumulated.surface.area.of.cells.in.a.colony.µm²)
size.dat$Colony.surface.area.µm²..with.mucilage.=as.numeric(size.dat$Colony.surface.area.µm²..with.mucilage.)
head(size.dat)
mfg.traits.dbase$large.MFG
# mfg.traits.dbase<-phyto.convert.df(mfg.traits.dbase)
mfg.traits.dbase$large.MFG=NA
mfg.traits.dbase$large.MFG[grep(c('1|6|8'),mfg.traits.dbase$MFG)]='L'
mfg.traits.dbase$large.MFG[grep(c('2|7|9'),mfg.traits.dbase$MFG)]='S'
mfg.traits.dbase$large.MFG
size.vars=names(mfg.traits.dbase)[c(7,15,16,17,9,18,22,23,24)]
size.dat=mfg.traits.dbase[,size.vars]
sapply(size.dat,function(x) sum(is.na(x)))
size.dat$Colony.surface.area.µm²..with.mucilage.[is.na(size.dat$Colony.surface.area.µm²..with.mucilage.)]=size.dat$Cell.surface.area.µm²[is.na(size.dat$Colony.surface.area.µm²..with.mucilage.)]
size.dat$Colony.biovolume.µm3..without.mucilage.[is.na(size.dat$Colony.biovolume.µm3..without.mucilage.)]=size.dat$Cell.biovolume.µm3[is.na(size.dat$Colony.biovolume.µm3..without.mucilage.)]
size.dat$large.MFG=mfg.traits.dbase$large.MFG
size.dat$large.MFG=ifelse(size.dat$large.MFG=='L',1,0)
size.dat$Nano.microphytoplankton=ifelse(size.dat$Nano.microphytoplankton=='Nanophytoplankton',0,1)
size.dat$Colonial[size.dat$Number.of.cells.per.colony>=2]=1
size.dat$Cell.surface.area.µm²=as.numeric(size.dat$Cell.surface.area.µm²)
size.dat$Cumulated.surface.area.of.cells.in.a.colony.µm²=as.numeric(size.dat$Cumulated.surface.area.of.cells.in.a.colony.µm²)
size.dat$Colony.surface.area.µm²..with.mucilage.=as.numeric(size.dat$Colony.surface.area.µm²..with.mucilage.)
size.dat
library(randomForest)
size.dat<-na.omit(size.dat)
size.rf<-randomForest(large.MFG~.,data=size.dat,family='binomial')
varImpPlot(size.rf)
partialPlot(size.rf,pred.data = size.dat,x.var = 'Colonial')
mld.pp=partialPlot(size.rf,pred.data = size.dat,x.var = 'Maximal.length.of.the.algal.object')
size.ctree<-ctree(large.MFG~.,data=size.dat)
size.dat$ctree.node=predict(size.ctree,size.dat,type='node')
plot(size.ctree)
library(party)
library(party)
size.ctree<-ctree(large.MFG~.,data=size.dat)
size.dat$ctree.node=predict(size.ctree,size.dat,type='node')
plot(size.ctree)
?ctree
table(mfg.traits.dbase$MFG==mfg.traits.dbase$trait.MFG)
prod=diag(confusion.mat)/apply(confusion.mat,1,sum)
users=diag(confusion.mat)/apply(confusion.mat,2,sum)
mean(na.omit(prod))
mean(na.omit(users))
table(mfg.traits.dbase$MFG==mfg.traits.dbase$trait.MFG)
table(mfg.traits.dbase$MFG==mfg.traits.dbase$trait.MFG)/sum(632,190)
source('species_mfg_convert.r')
?pvclust
library(pvclust)
new.mfg.library=mfg.traits.dbase[is.na(mfg.traits.dbase$MFG),]
head(new.mfg.library)
new.mfg.library=mfg.traits.dbase[is.na(mfg.traits.dbase$MFG)&!is.na(mfg.traits.dbase$trait.MFG),]
head(new.mfg.library)
table(is.na(mfg.traits.dbase$MFG))
239/(901+239)
239-188
51/1140
new.mfg.library=mfg.traits.dbase[is.na(mfg.traits.dbase$MFG)&!is.na(mfg.traits.dbase$trait.MFG),]
new.mfg.library=subset(new.mfg.library,select=c('genus','species','MFG'))
head(new.mfg.library)
new.mfg.library=mfg.traits.dbase[is.na(mfg.traits.dbase$MFG)&!is.na(mfg.traits.dbase$trait.MFG),]
new.mfg.library=subset(new.mfg.library,select=c('genus','species','trait.MFG'))
write.csv(new.mfg.library,'../mee ms/newsize_mfg_library_add.csv')
write.csv(species.mfg.library,'../mee ms/old_mfg_library.csv')
newspecies.mfg.library=read.csv('../mee ms/expanded_mfg_library.csv')
save('sppMFG_new.rda')
save(newspecies.mfg.library,'sppMFG_new.rda')
save(newspecies.mfg.library,file='sppMFG_new.rda')
save(newspecies.mfg.library,file='sppMFG_new.rda')
species.mfg.library=newspecies.mfg.library
names(mfg.traits.dbase)
?metaMDS
mfg.traits.forord=subset(mfg.traits.dbase,select=c('Heterotrophic','Mixotrophic','Autotrophic','Filament','Colonial','Gelatinous','aerotopes','Class','Order','Centric'))
mfg.ord=metaMDS(mfg.traits.forord)
mfg.traits.forord$Large=ifelse(mfg.traits.dbase=='large',1,0)
mfg.traits.forord[is.na(mfg.traits.forord)]=0
mfg.traits.forord$Large=ifelse(mfg.traits.dbase=='large',1,0)
str(mfg.traits.forord)
mfg.traits.forord=subset(mfg.traits.dbase,select=c('Heterotrophic','Mixotrophic','Autotrophic','Filament','Colonial','Gelatinous','aerotopes','Centric'))
mfg.traits.forord[is.na(mfg.traits.forord)]=0
mfg.traits.forord
mfg.ord=metaMDS(mfg.traits.forord)
plot(mfg.ord)
test=pvclust(mfg.traits.forord)
plot(test)
?vegdist
traitdist=vegdist(mfg.traits.forord,method='gower')
test2<-hclust(traitdist)
plot(Test2)
plot(test2)
test3=pvclust(traitdist)
test3=pvclust(mfg.traits.forord,method='gower')
?pvclust
test3=pvclust(mfg.traits.forord,method.dist='gower')
test3=pvclust(mfg.traits.forord,method.hclust='gower')
head(mfg.traits.forord)
row.names(mfg.traits.forord)=mfg.traits.dbase$phyto_name
row.names(mfg.traits.forord)=paste(1:dim(mfg.traits.forord),mfg.traits.dbase$phyto_name)
row.names(mfg.traits.forord)=paste(1:dim(mfg.traits.forord),mfg.traits.dbase$phyto_name,sep='_')
row.names(mfg.traits.forord)=paste(1:dim(mfg.traits.forord)[1],mfg.traits.dbase$phyto_name,sep='_')
test2<-hclust(traitdist)
test3=pvclust(mfg.traits.forord,method='gower')
?pvclust
test2<-hclust(traitdist)
plot(test2)
?hclust
plot(test3,labels=mfg.traits.dbase$MFG)
plot(test2,labels=mfg.traits.dbase$MFG)
plot(test2,labels=mfg.traits.dbase$MFG,cex=.3)
table(is.na(mfg.traits.dbase$MFG))
mfg.csr
write.csv(mfg.csr,'../mee ms/mfg_csr_library.csv')
setwd('~/gleon/Geisha/phyto_package/GEISHA_phytoplankton_github_shared/')
orlane<-read.csv('orlaneTraitData_MFGcompare_original.csv',stringsAsFactors=F)
#the gelatinous column has been flipped so that 1 means gelatinous and 0 means no mucilage.
##now creating a vector of the trait names in orlanes dataset that correspond to arguments the traits.to.mfg function
orlane.trait.names=c('Motile','size','Colonial','Filament','Centric','Gelatinous','aerotopes','Class','Order')
orlane<-read.csv('orlaneTraitData_MFGcompare_gelFix.csv',stringsAsFactors=F)
orlane<-read.csv('orlaneTraitData_MFGcompare_gelFix.csv',stringsAsFactors=F)
#the gelatinous column has been flipped so that 1 means gelatinous and 0 means no mucilage.
##now creating a vector of the trait names in orlanes dataset that correspond to arguments the traits.to.mfg function
orlane.trait.names=c('Motile','size','Colonial','Filament','Centric','Gelatinous','aerotopes','Class','Order')
###########################################################33
#load the scripts
source('species_mfg_convert02042018.r')
source('mfg_csr.r')
source('mfg_to_traits.r')
source('traits_to_mfg02042018.r')
source('~/gleon/Geisha/phyto_package/GEISHA_phytoplankton_github_shared/mfg_to_traits.r', echo=TRUE)
source('~/gleon/Geisha/phyto_package/GEISHA_phytoplankton_github_shared/mfg_to_traits.r', echo=TRUE)
####example script demonstrating how to classify MFG and CSR using R scripts for the Geisha R package
####2/4/2018
rm(list=ls())
##first set the working directory
setwd('~/gleon/Geisha/phyto_package/GEISHA_phytoplankton_github_shared/')
##the directory should contain the following scripts:
##species_mfg_convert.r
##MFG_to_CSR.r
##traits_to_mfg.r
##mfg_to_traits.r
###you will also need the following R data objects:
##sppMFG.rda
##MFG_CSR.rda
###current versions of these scripts are labelled with date stamps
#####################################################################################
#load geneva dataset with species names and trait matrix
orlane<-read.csv('orlaneTraitData_MFGcompare_gelFix.csv',stringsAsFactors=F)
#the gelatinous column has been flipped so that 1 means gelatinous and 0 means no mucilage.
##now creating a vector of the trait names in orlanes dataset that correspond to arguments the traits.to.mfg function
orlane.trait.names=c('Motile','size','Colonial','Filament','Centric','Gelatinous','aerotopes','Class','Order')
###########################################################33
#load the scripts
source('species_mfg_convert02042018.r')
source('mfg_csr.r')
source('mfg_to_traits.r')
source('traits_to_mfg02042018.r')
load('MFG_CSR.rda')
load('sppMFG.rda')
#############################################################
##first, split the Taxa.name column into a genus and species
orlane<-genus.species.extract(orlane,'Taxa.name')
###now create MFG classification based on Nico's expert assessment
orlane<-phyto.convert.df(orlane)
##making second MFG column based on traits
##it will be called mfg.from.traits
orlane=trait.to.mfg.df(orlane,orlane.trait.names)
##making MFG number column associated with MFG.from.Nico
orlane$MFG.num=sapply(orlane$MFG,function(x) strsplit(x,split='-')[[1]][1])
##creating expert CSR classifications from Nico MFG classifications
orlane<-mfg.csr.convert.df(orlane,'MFG.num')
#comparisons
##missing classifications
dim(orlane)[1]
sum(is.na(orlane$mfg.from.traits))
sum(is.na(orlane$MFG))
table(orlane$mfg.from.traits==orlane$MFG)
###only about 60% agreement currently for lake geneva data.
###fraction caused by discrepancies in size
MFG.large=grep('L',orlane$MFG,ignore.case=F)
traits.large=grep('large',orlane$size)
MFG.small=grep('Small',orlane$MFG)
traits.small=grep('small',orlane$size)
##number of large MFG's listed as small in frederic's dataset
length(na.omit(match(MFG.large,traits.small)))
##number of small MFG's listed as large in frederic's dataset
length(na.omit(match(MFG.small,traits.large)))
#so 254 out of 371 mismatches between trait-based and expert classifications were associated with size.
###fraction caused by discrepancies in size
MFG.large=grep('L',orlane$MFG,ignore.case=F)
traits.large=grep('large',orlane$size)
MFG.small=grep('Small',orlane$MFG)
traits.small=grep('small',orlane$size)
#colonial mfgs
MFG.colonial=orlane$MFG.num %in% c('5a','5b','5c','5d','5e','6a1','6b1','10a','10b','10c','11a','11b','11c')
MFG.mismatch=orlane$MFG != orlane$mfg.from.traits
#species listed as unicellular in trait data
mfg.traits.unicell=orlane$Colonial==0
sum(MFG.colonial & mfg.traits.unicell)
##21 species listed as unicellular that nico classified as colonial
##discrepancies based on presence of mucilage
sum(na.omit(orlane$MFG=='11b-GelaChlor' & orlane$Gelatinous==0))
##52 mismatches due to gelatinous
head(orlane) ###view dataset with new columns