Análisis

Muestra de primeras lineas del dataset

ensayo

trat

hibrido

rep

l_tot

l_ch

l_med

l_gde

e_inf

e_med

e_sup

nlb_score

date

inoc

ddinoc

score

1 - 2023

DDD

MR

1

0

0

0

0

0

0

0

0

2023-03-08

2023-01-26

41 days

FALSE

1 - 2023

DDD

MR

2

0

0

0

0

0

0

0

0

2023-03-08

2023-01-26

41 days

FALSE

1 - 2023

DDD

MR

3

0

0

0

0

0

0

0

0

2023-03-08

2023-01-26

41 days

FALSE

1 - 2023

DDD

MR

4

0

0

0

0

0

0

0

0

2023-03-08

2023-01-26

41 days

FALSE

1 - 2023

DDD

MR

5

0

0

0

0

0

0

0

0

2023-03-08

2023-01-26

41 days

FALSE

1 - 2023

DDD

MR

6

2

0

0

2

2

0

0

0

2023-03-08

2023-01-26

41 days

FALSE

Tamaño de lesión

Mosaico de datos

datos originales: Primer eval sin el Bco, ultima todos 0

  • Calculo de AUC

2023

Analysis of Deviance Table (Type III tests)

Response: auc_manchas
                    LR Chisq Df Pr(>Chisq)    
trat                  80.794  3  < 2.2e-16 ***
lesion_size           27.524  2  1.055e-06 ***
hibrido                4.217  1    0.04001 *  
rep                    1.070  1    0.30097    
trat:lesion_size      47.012  6  1.861e-08 ***
trat:hibrido          47.423  3  2.826e-10 ***
lesion_size:hibrido   24.869  2  3.979e-06 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

trat:lesion_size

lesion_size = ch, hibrido = MR:
 trat response     SE  df asymp.LCL asymp.UCL .group
 JJJ    41.966 14.271 Inf    21.550    81.724  1    
 TTS     6.172  2.215 Inf     3.055    12.469   2   
 BLCO    1.609  0.667 Inf     0.714     3.627    3  
 DDD     0.572  0.278 Inf     0.221     1.484    3  

lesion_size = med, hibrido = MR:
 trat response     SE  df asymp.LCL asymp.UCL .group
 JJJ    17.054  5.839 Inf     8.717    33.365  1    
 BLCO    7.410  2.664 Inf     3.662    14.991  12   
 TTS     4.558  1.658 Inf     2.235     9.297   2   
 DDD     0.462  0.230 Inf     0.174     1.227    3  

lesion_size = gde, hibrido = MR:
 trat response     SE  df asymp.LCL asymp.UCL .group
 JJJ    12.993  4.533 Inf     6.558    25.743  1    
 DDD     0.496  0.246 Inf     0.187     1.313   2   
 BLCO    0.495  0.245 Inf     0.188     1.306   2   
 TTS     0.302  0.157 Inf     0.109     0.836   2   

lesion_size = ch, hibrido = S:
 trat response     SE  df asymp.LCL asymp.UCL .group
 TTS    92.794 31.830 Inf    47.375   181.759  1    
 JJJ    77.424 26.240 Inf    39.847   150.439  1    
 DDD     1.964  0.802 Inf     0.882     4.373   2   
 BLCO    0.590  0.281 Inf     0.232     1.499   2   

lesion_size = med, hibrido = S:
 trat response     SE  df asymp.LCL asymp.UCL .group
 TTS   292.301 99.899 Inf   149.596   571.139  1    
 JJJ   134.207 45.255 Inf    69.302   259.899  1    
 BLCO   11.587  4.106 Inf     5.785    23.206   2   
 DDD     6.762  2.508 Inf     3.268    13.990   2   

lesion_size = gde, hibrido = S:
 trat response     SE  df asymp.LCL asymp.UCL .group
 JJJ   251.303 85.711 Inf   128.790   490.359  1    
 TTS    47.663 17.007 Inf    23.684    95.919   2   
 DDD    17.831  6.429 Inf     8.796    36.149   2   
 BLCO    1.904  0.780 Inf     0.853     4.249    3  

Confidence level used: 0.95 
Intervals are back-transformed from the log scale 
P value adjustment: tukey method for comparing a family of 4 estimates 
Tests are performed on the log scale 
significance level used: alpha = 0.05 
NOTE: If two or more means share the same grouping symbol,
      then we cannot show them to be different.
      But we also did not show them to be the same. 

2024

Analysis of Deviance Table (Type III tests)

Response: auc_manchas
                         LR Chisq Df Pr(>Chisq)    
trat                       24.106  2  5.828e-06 ***
lesion_size               117.052  2  < 2.2e-16 ***
hibrido                    32.902  1  9.694e-09 ***
rep                         3.650  1    0.05606 .  
trat:lesion_size            5.904  4    0.20646    
trat:hibrido               20.772  2  3.087e-05 ***
lesion_size:hibrido         3.253  2    0.19660    
trat:lesion_size:hibrido    7.788  4    0.09968 .  
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

comparaciones multiples

lesion_size = ch, hibrido = MR:
 trat response     SE  df asymp.LCL asymp.UCL .group
 BLCO    17.42  4.534 Inf   10.4632        29  1    
 TTS     10.08  2.689 Inf    5.9721        17  1    
 DDD      2.20  0.700 Inf    1.1831         4   2   

lesion_size = med, hibrido = MR:
 trat response     SE  df asymp.LCL asymp.UCL .group
 TTS      0.73  0.307 Inf    0.3204         2  1    
 BLCO     0.24  0.153 Inf    0.0687         1  1    
 DDD      0.00  0.000 Inf    0.0000       Inf  1    

lesion_size = gde, hibrido = MR:
 trat response     SE  df asymp.LCL asymp.UCL .group
 DDD      0.00  0.000 Inf    0.0000       Inf  1    
 TTS      0.00  0.000 Inf    0.0000       Inf  1    
 BLCO     0.00  0.000 Inf    0.0000       Inf  1    

lesion_size = ch, hibrido = S:
 trat response     SE  df asymp.LCL asymp.UCL .group
 DDD    241.12 60.629 Inf  147.3011       395  1    
 TTS    178.86 45.016 Inf  109.2105       293  1    
 BLCO   164.77 41.485 Inf  100.5967       270  1    

lesion_size = med, hibrido = S:
 trat response     SE  df asymp.LCL asymp.UCL .group
 DDD    116.00 29.261 Inf   70.7523       190  1    
 TTS     16.76  4.368 Inf   10.0583        28   2   
 BLCO     8.50  2.292 Inf    5.0086        14   2   

lesion_size = gde, hibrido = S:
 trat response     SE  df asymp.LCL asymp.UCL .group
 DDD      5.83  1.620 Inf    3.3826        10  1    
 BLCO     0.00  0.000 Inf    0.0000       Inf  1    
 TTS      0.00  0.000 Inf    0.0000       Inf  1    

Confidence level used: 0.95 
Intervals are back-transformed from the log scale 
P value adjustment: tukey method for comparing a family of 3 estimates 
Tests are performed on the log scale 
significance level used: alpha = 0.05 
NOTE: If two or more means share the same grouping symbol,
      then we cannot show them to be different.
      But we also did not show them to be the same. 

Posición en el canopeo

2023

Analysis of Deviance Table (Type III tests)

Response: auc_manchas
                        LR Chisq Df Pr(>Chisq)    
trat                      42.897  3  2.588e-09 ***
lesion_pos                20.755  2  3.113e-05 ***
hibrido                    0.173  1    0.67777    
rep                        1.947  1    0.16293    
trat:lesion_pos            6.007  6    0.42238    
trat:hibrido               7.831  3    0.04963 *  
lesion_pos:hibrido         5.246  2    0.07257 .  
trat:lesion_pos:hibrido    1.112  6    0.98100    
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

trat:lesion_size:hibrido

lesion_pos = inf, hibrido = MR:
 trat response      SE  df asymp.LCL asymp.UCL .group
 JJJ    91.368  38.922 Inf    39.645       211  1    
 TTS     8.236   3.641 Inf     3.463        20   2   
 BLCO    3.934   1.812 Inf     1.595        10   2   
 DDD     2.913   1.377 Inf     1.154         7   2   

lesion_pos = med, hibrido = MR:
 trat response      SE  df asymp.LCL asymp.UCL .group
 JJJ    17.983   7.779 Inf     7.703        42  1    
 TTS     7.614   3.377 Inf     3.193        18  1    
 BLCO    5.274   2.382 Inf     2.176        13  12   
 DDD     0.859   0.490 Inf     0.280         3   2   

lesion_pos = sup, hibrido = MR:
 trat response      SE  df asymp.LCL asymp.UCL .group
 BLCO    0.000   0.000 Inf     0.000       Inf  1    
 DDD     0.000   0.000 Inf     0.000       Inf  1    
 JJJ     0.000   0.000 Inf     0.000       Inf  1    
 TTS     0.000   0.000 Inf     0.000       Inf  1    

lesion_pos = inf, hibrido = S:
 trat response      SE  df asymp.LCL asymp.UCL .group
 JJJ   364.728 154.926 Inf   158.637       839  1    
 TTS   116.003  49.376 Inf    50.369       267  12   
 DDD    26.216  11.273 Inf    11.286        61   2   
 BLCO    5.143   2.326 Inf     2.119        12    3  

lesion_pos = med, hibrido = S:
 trat response      SE  df asymp.LCL asymp.UCL .group
 TTS   345.326 146.692 Inf   150.191       794  1    
 JJJ   212.636  90.384 Inf    92.432       489  1    
 DDD     9.062   3.992 Inf     3.822        21   2   
 BLCO    8.438   3.727 Inf     3.551        20   2   

lesion_pos = sup, hibrido = S:
 trat response      SE  df asymp.LCL asymp.UCL .group
 TTS    52.602  22.419 Inf    22.815       121  1    
 JJJ    18.240   7.815 Inf     7.876        42  1    
 BLCO    0.966   0.535 Inf     0.326         3   2   
 DDD     0.000   0.000 Inf     0.000       Inf  12   

Confidence level used: 0.95 
Intervals are back-transformed from the log scale 
P value adjustment: tukey method for comparing a family of 4 estimates 
Tests are performed on the log scale 
significance level used: alpha = 0.05 
NOTE: If two or more means share the same grouping symbol,
      then we cannot show them to be different.
      But we also did not show them to be the same. 

2024

Analysis of Deviance Table (Type III tests)

Response: auc_manchas
                        LR Chisq Df Pr(>Chisq)    
trat                      42.897  3  2.588e-09 ***
lesion_pos                20.755  2  3.113e-05 ***
hibrido                    0.173  1    0.67777    
rep                        1.947  1    0.16293    
trat:lesion_pos            6.007  6    0.42238    
trat:hibrido               7.831  3    0.04963 *  
lesion_pos:hibrido         5.246  2    0.07257 .  
trat:lesion_pos:hibrido    1.112  6    0.98100    
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

trat:lesion_size:hibrido

lesion_pos = inf, hibrido = MR:
 trat response     SE  df asymp.LCL asymp.UCL .group
 BLCO    12.41  3.101 Inf     7.601        20  1    
 TTS      5.74  1.523 Inf     3.416        10  1    
 DDD      2.22  0.678 Inf     1.220         4   2   

lesion_pos = med, hibrido = MR:
 trat response     SE  df asymp.LCL asymp.UCL .group
 TTS      3.90  1.084 Inf     2.264         7  1    
 BLCO     3.77  1.052 Inf     2.182         7  1    
 DDD      0.00  0.000 Inf     0.000       Inf  1    

lesion_pos = sup, hibrido = MR:
 trat response     SE  df asymp.LCL asymp.UCL .group
 BLCO     1.47  0.492 Inf     0.760         3  1    
 TTS      1.23  0.433 Inf     0.619         2  1    
 DDD      0.00  0.000 Inf     0.000       Inf  1    

lesion_pos = inf, hibrido = S:
 trat response     SE  df asymp.LCL asymp.UCL .group
 DDD    203.90 48.318 Inf   128.151       324  1    
 BLCO    98.77 23.496 Inf    61.963       157  12   
 TTS     81.80 19.489 Inf    51.281       130   2   

lesion_pos = med, hibrido = S:
 trat response     SE  df asymp.LCL asymp.UCL .group
 DDD    132.54 31.469 Inf    83.227       211  1    
 TTS     70.73 16.874 Inf    44.308       113  1    
 BLCO    61.57 14.712 Inf    38.543        98  1    

lesion_pos = sup, hibrido = S:
 trat response     SE  df asymp.LCL asymp.UCL .group
 TTS     41.93 10.075 Inf    26.181        67  1    
 DDD     23.01  5.608 Inf    14.275        37  12   
 BLCO    12.81  3.197 Inf     7.856        21   2   

Confidence level used: 0.95 
Intervals are back-transformed from the log scale 
P value adjustment: tukey method for comparing a family of 3 estimates 
Tests are performed on the log scale 
significance level used: alpha = 0.05 
NOTE: If two or more means share the same grouping symbol,
      then we cannot show them to be different.
      But we also did not show them to be the same. 

Score

2023

trat = DDD:
 model term df1 df2 F.ratio  Chisq p.value
 hibrido      1 Inf   0.092  0.092  0.7620

trat = JJJ:
 model term df1 df2 F.ratio  Chisq p.value
 hibrido      1 Inf   7.913  7.913  0.0049

trat = TTS:
 model term df1 df2 F.ratio  Chisq p.value
 hibrido      1 Inf  18.167 18.167  <.0001
hibrido = MR:
 model term df1 df2 F.ratio  Chisq p.value
 trat         2 Inf   1.048  2.096  0.3507

hibrido = S:
 model term df1 df2 F.ratio  Chisq p.value
 trat         2 Inf   8.261 16.522  0.0003
hibrido = MR:
 trat emmean    SE  df asymp.LCL asymp.UCL .group
 JJJ  -2.742 0.861 Inf    -4.429     -1.06  1    
 DDD  -3.948 0.922 Inf    -5.756     -2.14  1    
 TTS  -4.134 0.965 Inf    -6.025     -2.24  1    

hibrido = S:
 trat emmean    SE  df asymp.LCL asymp.UCL .group
 TTS   1.848 0.803 Inf     0.274      3.42  1    
 JJJ   0.455 0.732 Inf    -0.979      1.89  1    
 DDD  -3.650 0.937 Inf    -5.487     -1.81   2   

Confidence level used: 0.95 
P value adjustment: tukey method for comparing a family of 3 estimates 
significance level used: alpha = 0.05 
NOTE: If two or more means share the same grouping symbol,
      then we cannot show them to be different.
      But we also did not show them to be the same. 

2024

hibrido = MR:
 model term df1 df2 F.ratio Chisq p.value
 trat         1 Inf   0.000  0.00  0.9984

hibrido = S:
 model term df1 df2 F.ratio Chisq p.value
 trat         1 Inf   0.920  0.92  0.3376
hibrido = MR:
 trat emmean    SE  df asymp.LCL asymp.UCL .group
 DDD  -5.396 1.393 Inf    -8.126     -2.67  1    
 TTS  -5.399 1.394 Inf    -8.131     -2.67  1    

hibrido = S:
 trat emmean    SE  df asymp.LCL asymp.UCL .group
 TTS   0.992 0.633 Inf    -0.247      2.23  1    
 DDD   0.255 0.564 Inf    -0.850      1.36  1    

Confidence level used: 0.95 
significance level used: alpha = 0.05 
NOTE: If two or more means share the same grouping symbol,
      then we cannot show them to be different.
      But we also did not show them to be the same.