Danie:Notebook/Haplotyping/Genotyping
Jump to navigation
Jump to search
Genotyping
This page focuses on my work regarding genotyping the different data sets from the haplotyping project. Genotyping is, in a sense, the first step in haplotyping, because you need to be able to identify heterozygous SNPs. Most programs, like HAPCUT, require a vcf that already states the locations of the heterozygous SNPs before running.
Analyzing Genotyping Abilities of Different Data Sets
In the final step, I use the following MATLAB code to create the ROC curve. You can view the master file as media on the wiki.
The steps are:
- Read in the infofile, which is a file that lists all the files you'd like to use for the analysis. The files may also have a tag, which is used for labeling. One file must be labeled "true" in order to analyze everything against a true VCF
- Next loads in position data for true file, and position and quality data for the test data
- Compares true positions against test positions, overlaps with the true data are considered "true" and all others are "false"
- False discovery rate and sensitivity are calculated using the "false" and "true" data sets, respectively.
Individual Data Sets
BAC
BAC was able to be analyzed by freebayes using its standard bam files. Since each bam was sorted by index, not chromosome, I used every bam file, extracting each chromosome for analysis individually. script