#setwd("~/Dropbox/MyR/consult/ISA/tese/aa estabilidade")# lab
#setwd("C:/Users/juanchi/Dropbox/MyR/consult/ISA/tese/aa estabilidade") #dell
load("~/Dropbox/MyR/consult/ISA/tese/aa estabilidade/scientia.RData")
#dat = read.csv("est_taxa.csv", dec=",", header=T, sep="\t", check.names=FALSE)
#dat_ged = read.csv("aval_estab.csv", dec=",", header=T, sep="\t", check.names=FALSE, na.strings=".")  
library(ggplot2);library(lme4)

## Loading required package: Matrix

knitr::opts_chunk$set(message=FALSE, warning=FALSE, echo=FALSE, fig.width = 6, fig.height = 4)

Micelial growth

(growth rate in vitro)

Interaction model ~ pooled data

## 
##  F test to compare two variances
## 
## data:  lm(ca ~ transf * iso, data = subset(dat, exp == 1)) and lm(ca ~ transf * iso, data = subset(dat, exp == 2))
## F = 0.49913, num df = 27, denom df = 27, p-value = 0.0765
## alternative hypothesis: true ratio of variances is not equal to 1
## 95 percent confidence interval:
##  0.2309767 1.0785872
## sample estimates:
## ratio of variances 
##          0.4991277

## Analysis of Variance Table
## 
## Response: ca
##            Df Sum Sq Mean Sq  F value Pr(>F)    
## transf      1 0.0013 0.00130   0.3084 0.5807    
## iso         2 4.3079 2.15393 509.7751 <2e-16 ***
## transf:iso  2 0.0066 0.00328   0.7770 0.4644    
## Residuals  60 0.2535 0.00423                    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Por isolado

## Analysis of Variance Table
## 
## Response: ca
##           Df   Sum Sq   Mean Sq F value Pr(>F)
## transf     1 0.000336 0.0003358  0.0939 0.7624
## Residuals 20 0.071477 0.0035739

## Analysis of Variance Table
## 
## Response: ca
##           Df Sum Sq Mean Sq F value Pr(>F)
## transf     1 0.0075 0.00750  2.1307 0.1599
## Residuals 20 0.0704 0.00352

## Analysis of Variance Table
## 
## Response: ca
##           Df   Sum Sq   Mean Sq F value Pr(>F)
## transf     1 0.000033 0.0000331  0.0059 0.9394
## Residuals 20 0.111638 0.0055819

T-test

## 
##  F test to compare two variances
## 
## data:  subset(dat, iso == "SP09839" & transf == 0)$ca and subset(dat, iso == "SP09839" & transf == 10)$ca
## F = 9.1905, num df = 1, denom df = 1, p-value = 0.4057
## alternative hypothesis: true ratio of variances is not equal to 1
## 95 percent confidence interval:
##  1.418744e-02 5.953486e+03
## sample estimates:
## ratio of variances 
##           9.190471

## 
##  Two Sample t-test
## 
## data:  subset(dat, iso == "SP09839" & transf == 0)$ca and subset(dat, iso == "SP09839" & transf == 10)$ca
## t = 1.706, df = 2, p-value = 0.2301
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -0.1546827  0.3579327
## sample estimates:
## mean of x mean of y 
## 0.6145714 0.5129464

## 
##  F test to compare two variances
## 
## data:  subset(dat, iso == "SP08345" & transf == 0)$ca and subset(dat, iso == "SP08345" & transf == 10)$ca
## F = 0.70968, num df = 1, denom df = 1, p-value = 0.8914
## alternative hypothesis: true ratio of variances is not equal to 1
## 95 percent confidence interval:
##  1.09554e-03 4.59722e+02
## sample estimates:
## ratio of variances 
##          0.7096786

## 
##  Two Sample t-test
## 
## data:  subset(dat, iso == "SP08345" & transf == 0)$ca and subset(dat, iso == "SP08345" & transf == 10)$ca
## t = -0.80651, df = 2, p-value = 0.5046
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -0.1574681  0.1077538
## sample estimates:
## mean of x mean of y 
## 0.4780000 0.5028571

---

Germina

Interaction model ~ pooled data

## 
##  F test to compare two variances
## 
## data:  lm(agerm ~ transf * iso, data = subset(dat_ged, exp == 1)) and lm(agerm ~ transf * iso, data = subset(dat_ged, exp == 2))
## F = 0.82706, num df = 98, denom df = 89, p-value = 0.3583
## alternative hypothesis: true ratio of variances is not equal to 1
## 95 percent confidence interval:
##  0.548667 1.241656
## sample estimates:
## ratio of variances 
##          0.8270596

## Analysis of Variance Table
## 
## Response: agerm
##             Df Sum Sq Mean Sq F value    Pr(>F)    
## transf       1 1.0728 1.07285 46.9995 9.331e-11 ***
## iso          2 0.2354 0.11768  5.1553   0.00659 ** 
## transf:iso   2 0.0170 0.00850  0.3726   0.68947    
## Residuals  193 4.4056 0.02283                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Por isolado

## Analysis of Variance Table
## 
## Response: agerm
##           Df  Sum Sq Mean Sq F value    Pr(>F)    
## transf     1 0.36047 0.36047  19.718 3.328e-05 ***
## Residuals 69 1.26142 0.01828                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

## Analysis of Variance Table
## 
## Response: agerm
##           Df  Sum Sq Mean Sq F value    Pr(>F)    
## transf     1 0.52014 0.52014  28.216 1.157e-06 ***
## Residuals 72 1.32727 0.01843                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

## Analysis of Variance Table
## 
## Response: agerm
##           Df  Sum Sq Mean Sq F value  Pr(>F)  
## transf     1 0.23704 0.23704  6.7842 0.01196 *
## Residuals 52 1.81689 0.03494                  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

---

Esporula

A falta de significancia na interaçaõ implica que a queda da esporulação foi para os tres isolados igual. Pode dever-se a uma questao experimental ou da natureza da variavel.

## 
##  F test to compare two variances
## 
## data:  lm(sqrt(espor) ~ transf * iso, data = subset(dat_ged, exp ==  and lm(sqrt(espor) ~ transf * iso, data = subset(dat_ged, exp ==     1)) and     2))
## F = 0.88354, num df = 102, denom df = 105, p-value = 0.5307
## alternative hypothesis: true ratio of variances is not equal to 1
## 95 percent confidence interval:
##  0.6000439 1.3023977
## sample estimates:
## ratio of variances 
##          0.8835359

## Analysis of Variance Table
## 
## Response: sqrt(espor)
##             Df Sum Sq Mean Sq F value  Pr(>F)  
## transf       1  1.022 1.02238  5.2544 0.02287 *
## iso          2  0.683 0.34159  1.7555 0.17530  
## transf:iso   2  0.272 0.13607  0.6993 0.49805  
## Residuals  213 41.445 0.19458                  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Por isolado

## Analysis of Variance Table
## 
## Response: sqrt(espor)
##           Df  Sum Sq Mean Sq F value Pr(>F)
## transf     1  0.2488 0.24883  1.6044 0.2093
## Residuals 74 11.4770 0.15510

## Analysis of Variance Table
## 
## Response: sqrt(espor)
##           Df  Sum Sq Mean Sq F value Pr(>F)
## transf     1  0.1327 0.13271  0.6241  0.432
## Residuals 77 16.3738 0.21265

## Analysis of Variance Table
## 
## Response: sqrt(espor)
##           Df  Sum Sq Mean Sq F value  Pr(>F)  
## transf     1  0.9213 0.92131   4.202 0.04461 *
## Residuals 62 13.5939 0.21926                  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

---

Fung4

Effect of transfers

## 
##  F test to compare two variances
## 
## data:  lm(fung4 ~ transf * iso, data = subset(dres, exp == 1)) and lm(fung4 ~ transf * iso, data = subset(dres, exp == 2))
## F = 0.65734, num df = 76, denom df = 76, p-value = 0.06934
## alternative hypothesis: true ratio of variances is not equal to 1
## 95 percent confidence interval:
##  0.4178698 1.0340553
## sample estimates:
## ratio of variances 
##          0.6573435

## Analysis of Variance Table
## 
## Response: fung4
##             Df Sum Sq  Mean Sq F value  Pr(>F)  
## transf       1 0.1931 0.193067  5.6210 0.01897 *
## iso          1 0.0150 0.015016  0.4372 0.50947  
## transf:iso   1 0.0411 0.041102  1.1967 0.27568  
## Residuals  156 5.3582 0.034347                  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Por isolado

## Analysis of Variance Table
## 
## Response: fung4
##           Df  Sum Sq  Mean Sq F value  Pr(>F)  
## transf     1 0.20616 0.206165    6.41 0.01336 *
## Residuals 78 2.50871 0.032163                  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

## Analysis of Variance Table
## 
## Response: fung4
##           Df Sum Sq  Mean Sq F value Pr(>F)
## transf     1 0.0280 0.028004  0.7666  0.384
## Residuals 78 2.8495 0.036532

T-test

##       iso transf exp        ca
## 1 SP09839      0   1 0.5580000
## 2 PR09638      0   1 0.9000000
## 3 SP08345      0   1 0.4581429
## 4 SP09839      1   1 0.5728571
## 5 PR09638      1   1 1.1139286
## 6 SP08345      1   1 0.4128571

## 
##  F test to compare two variances
## 
## data:  subset(dres, iso == "SP09839" & transf == 3)$fung4 and subset(dres, iso == "SP09839" & transf == 6)$fung4
## F = 0.40549, num df = 19, denom df = 19, p-value = 0.05607
## alternative hypothesis: true ratio of variances is not equal to 1
## 95 percent confidence interval:
##  0.160499 1.024457
## sample estimates:
## ratio of variances 
##          0.4054927

## 
##  Two Sample t-test
## 
## data:  subset(dres, iso == "SP09839" & transf == 3)$fung4 and subset(dres, iso == "SP09839" & transf == 6)$fung4
## t = -4.2364, df = 38, p-value = 0.0001392
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -0.3177399 -0.1122601
## sample estimates:
## mean of x mean of y 
##    0.6275    0.8425

## 
##  F test to compare two variances
## 
## data:  subset(dres, iso == "SP08345" & transf == 3)$fung4 and subset(dres, iso == "SP08345" & transf == 6)$fung4
## F = 0.19713, num df = 19, denom df = 19, p-value = 0.0008766
## alternative hypothesis: true ratio of variances is not equal to 1
## 95 percent confidence interval:
##  0.07802671 0.49804092
## sample estimates:
## ratio of variances 
##          0.1971307

## 
##  Welch Two Sample t-test
## 
## data:  subset(dres, iso == "SP08345" & transf == 3)$fung4 and subset(dres, iso == "SP08345" & transf == 6)$fung4
## t = -3.4604, df = 26.211, p-value = 0.001861
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -0.33071028 -0.08428972
## sample estimates:
## mean of x mean of y 
##    0.6000    0.8075

Resumo da analise de regress?o:

Efeito das Transferencias

H0: all transfers lead to the same effect

Isolates

H0: that all isolates are indistinguishable

Interaction isolates x transfer

H0: difference between isolates is consistent at all times

Variavel	transfer	Isolado	Interaction	? t-test: 0 vs 10
MG	0.893	<2e-16	0.464
Germina	1.7e-10	0.006	0.689
Esporula	0.023	0.174	0.498
MGinDD	0.115	0.666	0.283	p = 0.219 / 0.694

? SP09839 / SP08345