This is a mirror of agdamsbo/daDoctoR on GitHub
Go to file
2018-10-03 10:25:32 +02:00
man Still adding 2018-10-03 08:42:37 +02:00
R Fix? 2018-10-03 10:25:32 +02:00
.gitignore Updating 2018-10-02 21:19:32 +02:00
.Rbuildignore Updating 2018-10-02 21:19:32 +02:00
daDoctoR.Rproj Updating 2018-10-02 21:19:32 +02:00
DESCRIPTION Add files via upload 2018-10-02 21:07:43 +02:00
NAMESPACE Add files via upload 2018-10-02 21:07:43 +02:00
README.md Update README.md 2018-10-02 21:20:55 +02:00

Intro

I try my best to share my basic workarounds in R to more effectively do statistical research. Feel free to be inspired or comment.

Further research

In need of a suitable function to perform the chi-squared test of Hardy-Weinberg-equillibrium in my study poppulation, I ended up writing my own. It also contains a few summarise functions. This is actually the function I am most proud of, as it represents an actual universal test for both bi- and triallelic sustems in non-sexcromosome genes.

Genotype distribution testing

  • hwe_allele.R -- requires input in the form of two vectors with alleles listed
  • hwe_geno.R -- requires input as numbers of each genotype (mm, mn, nn for biallelic systems and mm, mn, nn, mo, no, oo for triallelic)
  • hwe.sum.R -- summarising tests for genotypes grouped by af factor. Performs HWE test for each group and returns neatly formatted distribution for easy copy-pasting to print. No comparisons of groups. Use the oddsratio or chisq.test.

Formatting large data frames

  • col_fact.R -- formatting columns as factor for names containing text elements of a vector provided. Labels or levels can be provided.
  • col_num.R -- formatting columns as numeric for names containing text elements of a vector provided.

Bivariate logistic regression analyses

  • rep_glm.R -- for a stepwise gating regression approach this provides several bivariate logistic regression analyses for columns of a dataframe specified by af vector of the format c(). Use the dput() to obtain names of dataframe in correct format.
  • cie_test.R -- Analysis of change in estimate approach with specified cut set at 10 % as standard.

Research year

This is my first attempt at creating something usefull for openly sharing. In my work as a research year student through Aarhus University, Denmark at Aarhus University Hospital, Denmark, I've been introduced to working in the programming language R. During this work I've come accross a few problems apparently unique for the work with data from the Danish medical databases. To analyse the data in R I've had to rewrite some commands and to write my own, so I thought that I should share them here for others to take advantage of. Nothing fancy, but hopefully you can skip af few steps on your way through data analysis. These functions are forked from my original uploads as drgammelgaard.

For a start I've uploaded five commands for extracting data from cpr-numbers:

  • age_calc_function.R
  • cpr_check_function.R
  • cpr_sex_function.R
  • date_convert_function.R
  • dob_extract_cpr_function.R

All of these commands are written to work with the Danish Central Person Registry (CPR) numbers of the format ddmmyy-xxxx.