1. Method

对于正态分布的两组数据的均值比较使用Student’s t检验，多组正态数据的均值比较使用方差检验，对于非正态数据的比较使用Mann-Whitney U 检验，多组非正态数据的比较使用Kruskal-Wallis rank sum检验，两组或多组数据样本比率比较使用卡方检验。所有数据统计和数据可视化使用R 语言版本3.6.3实现。以生存时间中位数为分界点，将样本分为转阴时间长组和短组。以转阴时间长短分组为因变量，以性别，年龄，诊断，民族，是否接种疫苗，是否加强针，有无并发症为自变量，建立logistic 回归模型，使用stepwise方法进行变量选择。

2. Reading Dataset

3. Summary

4. Summary by 性别

5. Summary by diagnosis group

6. Summary by age group

7. Summary by vaccine

8. Summary by 是否接种第三剂加强针

9. CT变化趋势 4.22－5.17

## 04220423042404250426042704280429043005010502050305040505050605070508050905100511051205130514051505160517

10. 研究不同转阴时间区间 <10天, 10－20天，>20天的人的CT值的差异。以第一次测得CT值为day1

(0,10] (10,20] (20,27] 879 667 36

N.CT point of samples with 双阴时间>20

png 2

Min. 1st Qu. Median Mean 3rd Qu. Max. 13.31 28.31 32.17 34.34 45.00 45.00

O.CT point of samples with 双阴时间>20

png 2

Min. 1st Qu. Median Mean 3rd Qu. Max. 11.06 29.05 33.07 35.33 45.00 45.00

612523199501142719 这个人按照提供文件转阴时间<=10, 但是从4.24-5.10共14天，均 有CT值

N.CT point of samples with 双阴时间<=10

png 2

Min. 1st Qu. Median Mean 3rd Qu. Max. 12.57 29.41 35.05 36.33 45.00 45.00

O.CT point of samples with 双阴时间<=10

png 2

Min.  1st Qu.   Median     Mean  3rd Qu.     Max.

0.08933 29.74679 37.01500 36.88297 45.00000 45.00000

N.CT wilcox rank test between 双阴时间<=10 和 >20

Wilcoxon rank sum test with continuity correction

data: N.CT_all_less10_ml\(N.CT and N.CT_all_large20_ml\)N.CT W = 1545572, p-value = 1.499e-08 alternative hypothesis: true location shift is not equal to 0

O.CT wilcox rank test between 双阴时间<=10 和 >20

Wilcoxon rank sum test with continuity correction

data: O.CT_all_less10_ml\(O.CT and O.CT_all_large20_ml\)O.CT W = 1282990, p-value = 1.75e-05 alternative hypothesis: true location shift is not equal to 0

N.CT point of samples with 双阴时间10 to 20

png 2

Min. 1st Qu. Median Mean 3rd Qu. Max. 3.071 28.355 32.973 34.781 45.000 45.000

O.CT point of samples with 双阴时间10-20

png 2

Min. 1st Qu. Median Mean 3rd Qu. Max. 1.37 28.94 33.78 35.60 45.00 45.00

N.CT kruskal.test rank test between 双阴时间<=10 10-20 和 >20

Kruskal-Wallis rank sum test

data: list(N.CT_all_less10_ml\(N.CT, N.CT_all_large20_ml\)N.CT, N.CT_all_10to20_ml$N.CT) Kruskal-Wallis chi-squared = 86.704, df = 2, p-value < 2.2e-16

O.CT kruskal.test rank test between 双阴时间<=10 10-20 和 >20

Kruskal-Wallis rank sum test

data: list(O.CT_all_less10_ml\(O.CT, O.CT_all_large20_ml\)O.CT, O.CT_all_10to20_ml$O.CT) Kruskal-Wallis chi-squared = 55.247, df = 2, p-value = 1.007e-12

11. 亚定点合并症

## 
## 女 男 
## 46 44

12. Logistic regression

[1] 10

Long Short 836 748

Full model:

Model after stepwise feature selction:

13. 8个复阳与其他样本比较(排除无诊断的，还有个别ID在总表里没有的人)