-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy path3_malsim_scenario1.R
491 lines (388 loc) · 22.1 KB
/
3_malsim_scenario1.R
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
##############################################
##
## Validation simulation for the cluster analysis of LLINEUP data
##
## Stage 1 we want to match as best as possible the
## trial data and see how closely we can recreate the
## cross-sectional survey results
## Functions from packages
# remotes::install_github("mrc-ide/cali")
## just needed this once to load in cali
# remotes::install_github("mrc-ide/malariasimulation")
library(malariasimulation)
library(cali)
# library(ggplot2)
## Data
test_data = read.csv("simulation-summary/scenario_1.csv",header=TRUE)
##########################################
##
## Function to run all sims first for calibrations to baseline prevalence
## assuming this is 1 month prior to the trial initiation
## Adjusting this to include a 2023 distribution running to 2026
## so we can compare to the 2 year net distribution campaign
## create the parameters list for the malaria model simulation:
cali_clusters_f = function(dat_row){
year <- 365
month <- 30
sim_length <- 15 * year ## initially just til 2020 as need extra resistance data!
## This is spanning Jan 2014 - Dec 2025
## Assume that all places have received nets at start of 2014
## then relative to jan 2017 (see uganda_what_if...)
## then as noted for 2020
## and finally we wish to see the difference for these places running forward from 2020-2023 (1 or 3 years)
human_population <- 20000
## This is calibrated to reflect the mosquito densities: raw data
starting_EIR <- 1
# dat_row = 1
simparams <- get_parameters(
list(
human_population = human_population,
# irs_correlation =
prevalence_rendering_min_ages = 2 * 365, ## Prev in 6 months to 14 years measured
prevalence_rendering_max_ages = 10 * 365,
model_seasonality = TRUE, ## Seasonality to match study site inputs [sites_13]
## match to estimates of seasonal trends from Pete Winskill GTS updates
g0 = test_data$seasonal_a0[dat_row],
g = c(test_data$seasonal_a1[dat_row], test_data$seasonal_a2[dat_row], test_data$seasonal_a3[dat_row]),
h = c(test_data$seasonal_b1[dat_row], test_data$seasonal_b2[dat_row], test_data$seasonal_b3[dat_row]),
Q0 = 0.72,
individual_mosquitoes = FALSE ## Update next
)
)
# set species
## match to trial data - gambiae sl to arabiensis: raw data
simparams <- set_species(simparams,
species=list(gamb_params, arab_params, fun_params),
proportions=c(0.4,
0.6, ## mosquitoes mostly became arabiensis dominated
0))
# set treatment
## an unknown and assuming from World Malaria Report country-wide averages
simparams <- set_drugs(simparams, list(AL_params, SP_AQ_params, DHA_PQP_params))
simparams <- set_clinical_treatment(simparams,
drug=1, ## In Uganda used for uncomplicated malaria
##https://www.cdc.gov/globalhealth/countries/uganda/pdf/uganda_factsheet.pdf
##Uganda population 2017 = 42862958
##ACT courses delivered 2017 = 27396300
## Estimated no of cases: 12140161
## Popn at risk: 44269584
time=c(100),
coverage=0.62) ## 27396300/44269584
simparams <- set_clinical_treatment(simparams,
drug=2, ## SP for severe...
time=c(100),
coverage=0.07) ## estimated pregnancy
## Set up bed nets
bednetparams <- simparams
## as done
bednet_events = data.frame(
timestep = c(0, 3, 6, 9) * year + c(0,
## raw data: net timing
test_data$days_after_jan_2017[dat_row],
test_data$days_after_jan_2017[dat_row],
test_data$days_after_jan_2017[dat_row]),
name=c("background",
"background",
"trial_nets",
"2020_nets")
)
# Will run over a loop so this is for either net
# these with be the pyr-ITNs/pob-ITNs depending on row of data
baseline_itn_use_estimates = mean(c(test_data$LLIN_USE_BASELINE[dat_row],
test_data$X3_and_over_net_use_baseline[dat_row]))
bednetparams_1 <- set_bednets(
bednetparams,
timesteps = bednet_events$timestep,
## raw data: net use prior and immediately after the campaign
coverages = c(baseline_itn_use_estimates, ## historic prior to RCT
baseline_itn_use_estimates, ## historic prior
test_data$sleeping_under_net_6m[dat_row], ## RCT (on deployment)
test_data$sleeping_under_net_6m[dat_row]), ## planned for 2020 - ** Assuming the distribution coverage matched the RCT estimate
## net attrition from raw data
retention = test_data$itn_leave_durMean[dat_row] * year, ## Keeping this as it was observed during RCT
## each row needs to show the efficacy parameter across years (and cols are diff mosquito) *note transposing so
## raw data: resistance status
dn0 = t(matrix(as.numeric(rep(c(test_data$dn0_med_2014[dat_row], test_data$dn0_med_2017[dat_row]),each=6)), nrow=3, ncol=4)),
rn = t(matrix(as.numeric(rep(c(test_data$rn0_med_2014[dat_row], test_data$rn0_med_2017[dat_row]),each=6)), nrow=3, ncol=4)),
rnm = matrix(c(.24, .24, .24, .24), nrow=4, ncol=3),
gamman = as.numeric(rep(c(test_data$gamman_med_2014[dat_row],test_data$gamman_med_2017[dat_row]),each=2) * 365)
)
##end
## assume the same people are getting nets each round
correlationsb1 <- get_correlation_parameters(bednetparams_1)
correlationsb1$inter_round_rho('bednets', 1)
simparams <- set_equilibrium(simparams, starting_EIR)
# Define target, here two prevalence measures:
target <- c(test_data$Prevalence_baseline_2_10_yrs[dat_row])
# Time points at which to match target
target_tt <- c(6*365+test_data$days_after_jan_2017[dat_row]-30)
## Run the simulations
output1 <- run_simulation(sim_length, bednetparams_1,correlationsb1)
set.seed(123)
out <- calibrate(parameters = bednetparams_1,
target = target,
target_tt = target_tt,
summary_function = summary_pfpr_2_10,
tolerance = 0.02,
interval = c(1, 500))##upper bound needs to be high enough so negative differences are not returned in uniroot
return(out$root)
}
# eir_est = numeric(nrow(test_data))
for(i in 1:nrow(test_data)){
eir_est[i] = cali_clusters_f(i)
}
test_data$eir_est = eir_est
write.csv(test_data,"C:/Users/esherrar/Documents/Rprojects/Key-RCT-metrics-LLINEUP/simulation-summary/scenario_1.csv")
plot(test_data$eir_est ~ test_data$Prevalence_baseline_2_10_yrs,
cex = 1+log(1+test_data$LLIN_USE_BASELINE),col=adegenet::transp("grey",0.6),pch=19)
##################################
##
## Simulate
test_data = read.csv("C:/Users/esherrar/Documents/Rprojects/Key-RCT-metrics-LLINEUP/simulation-summary/scenario_1.csv",header=TRUE)
uganda_1_f = function(dat_row){
year <- 365
month <- 30
sim_length <- 15 * year ## initially just til 2020 as need extra resistance data!
## This is spanning Jan 2014 - Dec 2025
## Assume that all places have received nets at start of 2014
## then relative to jan 2017 (see uganda_what_if...)
## then as noted for 2020
## and finally we wish to see the difference for these places running forward from 2020-2023 (1 or 3 years)
human_population <- 20000
## This is calibrated to reflect the mosquito densities: raw data
starting_EIR <- test_data$eir_est[dat_row]
simparams <- get_parameters(
list(
human_population = human_population,
# irs_correlation =
prevalence_rendering_min_ages = 2 * 365, ## Prev in 6 months to 14 years measured
prevalence_rendering_max_ages = 10 * 365,
model_seasonality = TRUE, ## Seasonality to match study site inputs [sites_13]
## match to estimates of seasonal trends from Pete Winskill GTS updates
g0 = test_data$seasonal_a0[dat_row],
g = c(test_data$seasonal_a1[dat_row], test_data$seasonal_a2[dat_row], test_data$seasonal_a3[dat_row]),
h = c(test_data$seasonal_b1[dat_row], test_data$seasonal_b2[dat_row], test_data$seasonal_b3[dat_row]),
Q0 = 0.72,
individual_mosquitoes = FALSE ## Update next
)
)
# set species
## match to trial data - gambiae sl to arabiensis: raw data
simparams <- set_species(simparams,
species=list(gamb_params, arab_params, fun_params),
proportions=c(0.4,
0.6, ## mosquitoes mostly became arabiensis dominated
0))
# set treatment
## an unknown and assuming from World Malaria Report country-wide averages
simparams <- set_drugs(simparams, list(AL_params, SP_AQ_params, DHA_PQP_params))
simparams <- set_clinical_treatment(simparams,
drug=1, ## In Uganda used for uncomplicated malaria
##https://www.cdc.gov/globalhealth/countries/uganda/pdf/uganda_factsheet.pdf
##Uganda population 2017 = 42862958
##ACT courses delivered 2017 = 27396300
## Estimated no of cases: 12140161
## Popn at risk: 44269584
time=c(100),
coverage=0.62) ## 27396300/44269584
simparams <- set_clinical_treatment(simparams,
drug=2, ## SP for severe...
time=c(100),
coverage=0.07) ## estimated pregnancy
## Set up bed nets
bednetparams <- simparams
## as done
bednet_events = data.frame(
timestep = c(0, 3, 6, 9) * year + c(0,
## raw data: net timing
test_data$days_after_jan_2017[dat_row],
test_data$days_after_jan_2017[dat_row],
test_data$days_after_jan_2017[dat_row]),
name=c("background",
"background",
"trial_nets",
"2020_nets")
)
# Will run over a loop so this is for either net
# these with be the pyr-ITNs/pob-ITNs depending on row of data
baseline_itn_use_estimates = mean(c(test_data$LLIN_USE_BASELINE[dat_row],
test_data$X3_and_over_net_use_baseline[dat_row]))
bednetparams_1 <- set_bednets(
bednetparams,
timesteps = bednet_events$timestep,
## raw data: net use prior and immediately after the campaign
coverages = c(baseline_itn_use_estimates, ## historic prior to RCT
baseline_itn_use_estimates, ## historic prior
test_data$sleeping_under_net_6m[dat_row], ## RCT (on deployment)
test_data$sleeping_under_net_6m[dat_row]), ## planned for 2020 - ** Assuming the distribution coverage matched the RCT estimate
## net attrition from raw data
retention = test_data$itn_leave_durMean[dat_row] * year, ## Keeping this as it was observed during RCT
## each row needs to show the efficacy parameter across years (and cols are diff mosquito) *note transposing so
## raw data: resistance status
dn0 = t(matrix(as.numeric(rep(c(test_data$dn0_med_2014[dat_row], test_data$dn0_med_2017[dat_row]),each=6)), nrow=3, ncol=4)),
rn = t(matrix(as.numeric(rep(c(test_data$rn0_med_2014[dat_row], test_data$rn0_med_2017[dat_row]),each=6)), nrow=3, ncol=4)),
rnm = matrix(c(.24, .24, .24, .24), nrow=4, ncol=3),
gamman = as.numeric(rep(c(test_data$gamman_med_2014[dat_row],test_data$gamman_med_2017[dat_row]),each=2) * 365)
)
##end
## assume the same people are getting nets each round
correlationsb1 <- get_correlation_parameters(bednetparams_1)
correlationsb1$inter_round_rho('bednets', 1)
bednetparams_1 <- set_equilibrium(bednetparams_1, starting_EIR)
## Run the simulations
output1 <- run_simulation(sim_length, bednetparams_1,correlationsb1)
output1$pv_730_3650 = output1$n_detect_730_3650/output1$n_730_3650
return(data.frame(timestep = output1$timestep,
prev_730_3650 = output1$pv_730_3650))
}
mod_sims_scenario_1 = expand.grid(
timestep = scenario_0$timestep
)
for(i in 1:nrow(test_data)){
mod_sims_scenario_1[,i+1] = uganda_1_f(i)[,2]
}
for(i in 1:nrow(test_data)){
colnames(mod_sims_scenario_1)[i+1] = paste0("arm",test_data$arm[i],"_prev")
}
write.csv(mod_sims_scenario_1, "simulation-summary/v2/mod_sims_scenario_1.csv")
########################################
##
## Starting here with presenting results
##
########################################
## simulations for prevalence
scenario_1 = read.csv("simulation-summary/mod_sims_scenario_1.csv",header=TRUE)
## input parameters for scenario
test_data = read.csv("C:/Users/esherrar/Documents/Rprojects/Key-RCT-metrics-LLINEUP/simulation-summary/scenario_1.csv",header=TRUE)
## looking across clusters
dat = read.csv("simulation-summary/scenario_2.csv",header=TRUE)
plot(scenario_1$arm1_prev ~ scenario_1$timestep,type="l",
ylab = "Prevalence in children 6m to 10yrs (%)",yaxt="n",ylim=c(0,0.5),
xlab = "Time in days",col="darkred",xlim=c(2300,3500)
)
cols=c("red","darkorange","darkblue","green")
year=365
dat_row=1
for(i in 1:4){
# Define target, here two prevalence measures:
target <- c(test_data$Prevalence_baseline_2_10_yrs[i])
# Time points at which to match target
target_tt <- c(6*365+test_data$days_after_jan_2017[i]-30)
points(target ~ target_tt,col=cols[i])
}
lines(mod_sims_scenario_0$arm1_prev ~ mod_sims_scenario_0$timestep,col="darkred")
lines(mod_sims_scenario_0$arm2_prev ~ mod_sims_scenario_0$timestep,col="orange")
lines(mod_sims_scenario_0$arm3_prev ~ mod_sims_scenario_0$timestep,col="blue")
lines(mod_sims_scenario_0$arm4_prev ~ mod_sims_scenario_0$timestep,col="darkgreen")
measured_moments= c(6*365+test_data$days_after_jan_2017)+c(365/2,365,365*1.5,365*2+30)
points(c(0.27,0.21,0.29,.109)~measured_moments,col="red",pch=19)
points(c(0.14,0.16,0.14,0.158)~measured_moments,col="orange",pch=19)
points(c(0.25,0.27,0.23,0.232)~measured_moments,col="blue",pch=19)
points(c(0.22,0.23,0.2,0.184)~measured_moments,col="green",pch=19)
lines(scenario_1$arm2_prev ~ scenario_1$timestep,col="orange")
lines(scenario_1$arm3_prev ~ scenario_1$timestep,col="blue")
lines(scenario_1$arm4_prev ~ scenario_1$timestep,col="darkgreen")
timestep = c(0, 3, 6, 9) * year + c(0,
## raw data: net timing
test_data$days_after_jan_2017[dat_row],
test_data$days_after_jan_2017[dat_row],
test_data$days_after_jan_2017[dat_row])
abline(v=timestep[3],lty=2)
points(dat$Prevalence_6m[dat$Net_Type == "Olyset Net"]~
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "Olyset Net"]+365/2),
col=adegenet::transp("red",0.2),pch=19)
points(dat$Prevalence_12m[dat$Net_Type == "Olyset Net"]~
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "Olyset Net"]+365),
col=adegenet::transp("red",0.2),pch=19)
points(dat$Prevalence_18m[dat$Net_Type == "Olyset Net"]~
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "Olyset Net"]+365+365/2),
col=adegenet::transp("red",0.2),pch=19)
points(dat$Prevalence_25m[dat$Net_Type == "Olyset Net"]~
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "Olyset Net"]+365*2+30),
col=adegenet::transp("red",0.2),pch=19)
dat_row=2
points(dat$Prevalence_6m[dat$Net_Type == "Olyset Plus"]~
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "Olyset Plus"]+365/2),
col=adegenet::transp("darkorange",0.2),pch=19)
points(dat$Prevalence_12m[dat$Net_Type == "Olyset Plus"]~
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "Olyset Plus"]+365),
col=adegenet::transp("darkorange",0.2),pch=19)
points(dat$Prevalence_18m[dat$Net_Type == "Olyset Plus"]~
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "Olyset Plus"]+365+365/2),
col=adegenet::transp("darkorange",0.2),pch=19)
points(dat$Prevalence_25m[dat$Net_Type == "Olyset Plus"]~
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "Olyset Plus"]+365*2+30),
col=adegenet::transp("darkorange",0.2),pch=19)
dat_row=3
points(dat$Prevalence_6m[dat$Net_Type == "PermaNet 2.0"]~
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "PermaNet 2.0"]+365/2),
col=adegenet::transp("darkblue",0.2),pch=19)
points(dat$Prevalence_12m[dat$Net_Type == "PermaNet 2.0"]~
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "PermaNet 2.0"]+365),
col=adegenet::transp("darkblue",0.2),pch=19)
points(dat$Prevalence_18m[dat$Net_Type == "PermaNet 2.0"]~
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "PermaNet 2.0"]+365+365/2),
col=adegenet::transp("darkblue",0.2),pch=19)
points(dat$Prevalence_25m[dat$Net_Type == "PermaNet 2.0"]~
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "PermaNet 2.0"]+365*2+30),
col=adegenet::transp("darkblue",0.2),pch=19)
dat_row=4
points(dat$Prevalence_6m[dat$Net_Type == "PermaNet 3.0"]~
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "PermaNet 3.0"]+365/2),
col=adegenet::transp("darkgreen",0.2),pch=19)
points(dat$Prevalence_12m[dat$Net_Type == "PermaNet 3.0"]~
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "PermaNet 3.0"]+365),
col=adegenet::transp("darkgreen",0.2),pch=19)
points(dat$Prevalence_18m[dat$Net_Type == "PermaNet 3.0"]~
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "PermaNet 3.0"]+365+365/2),
col=adegenet::transp("darkgreen",0.2),pch=19)
points(dat$Prevalence_25m[dat$Net_Type == "PermaNet 3.0"]~
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "PermaNet 3.0"]+365*2+30),
col=adegenet::transp("darkgreen",0.2),pch=19)
#######
##
## points versus time points
prev_obs = c(dat$Prevalence_6m[dat$Net_Type == "Olyset Net"],
dat$Prevalence_12m[dat$Net_Type == "Olyset Net"],
dat$Prevalence_18m[dat$Net_Type == "Olyset Net"],
dat$Prevalence_25m[dat$Net_Type == "Olyset Net"],
dat$Prevalence_6m[dat$Net_Type == "Olyset Plus"],
dat$Prevalence_12m[dat$Net_Type == "Olyset Plus"],
dat$Prevalence_18m[dat$Net_Type == "Olyset Plus"],
dat$Prevalence_25m[dat$Net_Type == "Olyset Plus"],
dat$Prevalence_6m[dat$Net_Type == "PermaNet 2.0"],
dat$Prevalence_12m[dat$Net_Type == "PermaNet 2.0"],
dat$Prevalence_18m[dat$Net_Type == "PermaNet 2.0"],
dat$Prevalence_25m[dat$Net_Type == "PermaNet 2.0"],
dat$Prevalence_6m[dat$Net_Type == "PermaNet 3.0"],
dat$Prevalence_12m[dat$Net_Type == "PermaNet 3.0"],
dat$Prevalence_18m[dat$Net_Type == "PermaNet 3.0"],
dat$Prevalence_25m[dat$Net_Type == "PermaNet 3.0"])
prev_mod = c(scenario_1$arm1_prev[c(c(6*365+dat$days_after_jan_2017[dat$Net_Type == "Olyset Net"]+365/2),
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "Olyset Net"]+365),
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "Olyset Net"]+365+365/2),
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "Olyset Net"]+365*2+30))],
scenario_1$arm2_prev[c(c(6*365+dat$days_after_jan_2017[dat$Net_Type == "Olyset Plus"]+365/2),
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "Olyset Plus"]+365),
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "Olyset Plus"]+365+365/2),
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "Olyset Plus"]+365*2+30))],
scenario_1$arm3_prev[c(c(6*365+dat$days_after_jan_2017[dat$Net_Type == "PermaNet 2.0"]+365/2),
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "PermaNet 2.0"]+365),
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "PermaNet 2.0"]+365+365/2),
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "PermaNet 2.0"]+365*2+30))],
scenario_1$arm4_prev[c(c(6*365+dat$days_after_jan_2017[dat$Net_Type == "PermaNet 3.0"]+365/2),
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "PermaNet 3.0"]+365),
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "PermaNet 3.0"]+365+365/2),
c(6*365+dat$days_after_jan_2017[dat$Net_Type == "PermaNet 3.0"]+365*2+30))])
plot(prev_obs ~ prev_mod, ylim=c(0,0.8),xlim=c(0,0.8),
col=adegenet::transp(rep(cols,each=4),0.1),pch=19)
abline(a=0,b=1,lty=2)
prev_obs_means = c(mean(prev_obs[1:52],na.rm=TRUE),
mean(prev_obs[53:128],na.rm=TRUE),
mean(prev_obs[129:288],na.rm=TRUE),
mean(prev_obs[289:416],na.rm=TRUE))
prev_mod_means = c(mean(prev_mod[1:52],na.rm=TRUE),
mean(prev_mod[53:128],na.rm=TRUE),
mean(prev_mod[129:288],na.rm=TRUE),
mean(prev_mod[289:416],na.rm=TRUE))
points(prev_obs_means ~ prev_mod_means,col=cols,pch=19,cex=1.2)