Angsd structure

History

1) A simple program to estimate the MAF based on files containing the counts of allleles (-doPhat). Used in (yu li 2009)

2) An implementation of (suyeon kim 2011). GL/error estimation based on counts of alleles along with proper GL MAF estimators and case/control association

3) To perform genomewide analysis a .soap file was reader was made and this was merged with 1) and 2). At this time the program was called 'dirtySoap'. Name was choosen, since it was a hack that used soap files.

4) A program called realSFS (nielsen 2012) to estimate the SFS was implemented and used text representation of glfv3. This was extended to read glfv2 and binary glv3.

5) The glfv3 reader from 4) was used to implement the score statistic of (skotte 2012).

6) The programs from 3) 4) 5) was merged. Program was still called dirtySoap

7) We extended the dirtySoap with the mpileup reading from samtools 1.17 and called the program angsd. This was implemented with the samtool beeing invoked from inside angsd.

8) We made a native bam reader (inspired and based on the samtools program (heng li)) along with the GATK,soapSNP GL model.

9) The theta estimator along with the tajima was then implemented (almost submitted).