整洁的多个方差分析

Question

我整理了数据，照样做了。

在我的办公室电脑上工作，不在我的家用电脑上工作。现在我得到：

Error in var(if (is.vector(x) || is.factor(x)) x else as.double(x), na.rm = na.rm) : 
  is.atomic(x) is not TRUE
In addition: Warning messages:
1: Data frame tidiers are deprecated and will be removed in an upcoming release of broom. 
2: In mean.default(X[[i]], ...) :
  argument is not numeric or logical: returning NA
3: In mean.default(X[[i]], ...) :
  argument is not numeric or logical: returning NA
4: In var(if (is.vector(x) || is.factor(x)) x else as.double(x), na.rm = na.rm) :
  NAs introduced by coercion

代码：

res = Raw.data %>% group_by(Gene) %>% 
  do(Model = aov(log2(FC) ~ Treatment, data=.))
tidy(res, Model)

问题出在 tidy(res, Model) 上，因为我使用 summary(res[[2]][[1]]) 等

得到了很好的统计数据

我真的很喜欢 tidy 在办公室工作时给我输出的方式。

数据：


   Number Treatment   Gene         FC
1       2   Control   mTOR 1.28999546
2       3   Control   mTOR 1.62429990
3       4   Control   mTOR 1.31081235
4      10   Control   mTOR 0.65558902
5      14   Control   mTOR 0.49470104
6      18   Control   mTOR 1.12261436
7       6 Treatment   mTOR 1.34369529
8       7 Treatment   mTOR 0.58483880
9       8 Treatment   mTOR 0.51403301
10      9 Treatment   mTOR 1.47711406
11     11 Treatment   mTOR 2.05220846
12     12 Treatment   mTOR 0.20960123
13     13 Treatment   mTOR 1.11679544
14     15 Treatment   mTOR 1.35787956
15     16 Treatment   mTOR 0.74617363
16     17 Treatment   mTOR 1.68791400
17     20 Treatment   mTOR 2.04683987
18     21 Treatment   mTOR 0.21358785
19     22 Treatment   mTOR 0.98309230
20     24 Treatment   mTOR 0.65445858
21     25 Treatment   mTOR 0.77690342
22     26 Treatment   mTOR 0.35951121
23      2   Control Raptor 0.82422904
24      3   Control Raptor 1.29085590
25      4   Control Raptor 0.55457309
26     10   Control Raptor 1.36949046
27     14   Control Raptor 0.90613140
28     18   Control Raptor 1.36573152
29      6 Treatment Raptor 0.40208821
30      7 Treatment Raptor 1.42850190
31      8 Treatment Raptor 0.47058962
32      9 Treatment Raptor 1.53576947
33     11 Treatment Raptor 2.08432767
34     12 Treatment Raptor 0.28285010
35     13 Treatment Raptor 1.28948941
36     15 Treatment Raptor 1.55241563
37     16 Treatment Raptor 1.03140971
38     17 Treatment Raptor 1.16624466
39     20 Treatment Raptor 0.25957711
40     21 Treatment Raptor 1.93043388
41     22 Treatment Raptor 2.71472997
42     24 Treatment Raptor 0.71381887
43     25 Treatment Raptor 1.47245399
44     26 Treatment Raptor 0.51014311
45      2   Control Rictor 0.76958681
46      3   Control Rictor 0.96147713
47      4   Control Rictor 0.89860880
48     10   Control Rictor 3.12117681
49     14   Control Rictor 0.32683138
50     18   Control Rictor 1.47431619
51      6 Treatment Rictor 0.02552013
52      7 Treatment Rictor 6.09665587
53      8 Treatment Rictor 1.00468371
54      9 Treatment Rictor 0.36000695
55     11 Treatment Rictor 1.54380977
56     12 Treatment Rictor 2.00068407
57     13 Treatment Rictor 0.40089656
58     15 Treatment Rictor 0.60702662
59     16 Treatment Rictor 1.06833716
60     17 Treatment Rictor 0.20441144
61     20 Treatment Rictor 1.18198006
62     21 Treatment Rictor 0.78789596
63     22 Treatment Rictor 0.04089405
64     24 Treatment Rictor 0.93654515
65     25 Treatment Rictor 0.80949506
66     26 Treatment Rictor 0.45670381
67      2   Control  mLST8 1.12922200
68      3   Control  mLST8 0.63262968
69      4   Control  mLST8 0.68963219
70     10   Control  mLST8 1.89325536
71     14   Control  mLST8 0.95387898
72     18   Control  mLST8 1.12396065
73      6 Treatment  mLST8 0.25950270
74      7 Treatment  mLST8 2.03655754
75      8 Treatment  mLST8 0.87489857
76      9 Treatment  mLST8 0.51938390
77     11 Treatment  mLST8 0.31708484
78     12 Treatment  mLST8 0.28315297
79     13 Treatment  mLST8 0.35406819
80     15 Treatment  mLST8 0.47686481
81     16 Treatment  mLST8 0.24946641
82     17 Treatment  mLST8 0.31349415
83     20 Treatment  mLST8 0.46643244
84     21 Treatment  mLST8 0.56343498
85     22 Treatment  mLST8 0.40902527
86     24 Treatment  mLST8 0.53124407
87     25 Treatment  mLST8 1.23766868
88     26 Treatment  mLST8 1.16274782

如有任何帮助，我们将不胜感激！

Answer 1

家用电脑和工作电脑的区别可能与 dplyr and/or broom?

的版本有关

尝试在嵌套数据的每一行上使用 nest_by（dplyr 版本 1.0.0）和运行而不是 group_by。使用 nest_by 将创建一个新的临时列表列 data。它类似于以前使用 nest 和 rowwise。该模型也需要在 list 中。

library(dplyr)
library(broom)

Raw.data %>%
  nest_by(Gene) %>%
  mutate(Model = list(aov(log2(FC) ~ Treatment, data = data))) %>%
  summarise(tidy(Model))

这应该允许您运行 aov 分别针对不同的基因并给出相似的输出。

输出

  Gene   term         df   sumsq meansq statistic p.value
  <chr>  <chr>     <dbl>   <dbl>  <dbl>     <dbl>   <dbl>
1 mLST8  Treatment     1  4.03   4.03      6.02    0.0235
2 mLST8  Residuals    20 13.4    0.670    NA      NA     
3 mTOR   Treatment     1  0.376  0.376     0.403   0.533 
4 mTOR   Residuals    20 18.7    0.934    NA      NA     
5 Raptor Treatment     1  0.0253 0.0253    0.0279  0.869 
6 Raptor Residuals    20 18.1    0.906    NA      NA     
7 Rictor Treatment     1  2.88   2.88      0.902   0.354 
8 Rictor Residuals    20 63.9    3.20     NA      NA

整洁的多个方差分析

Tidy multiple anovas

r

anova

tidy