Hi
I am interesting in alleles ancestral states, I noted Ensembl released EPO ancestral data 1 that based on the GRCh37 assembly of the human genome and compare with five of primates.
I found the data contain fasta and bed files
The README file said:
This directory contains the ancestral sequences for Homo sapiens (GRCh37).
The data have been extracted from the following alignment set: # Database: ensembl_compara_71@ens-staging2:3306 # MethodLinkSpeciesSet: 6 primates EPO (548)
In the EPO (Enredo-Pecan-Ortheus) pipeline, Ortheus infers ancestral states from the Pecan alignments. The confidence in the ancestral call is determined by comparing the call to the ancestor of the ancestral sequence as well as the 'sister' sequence of the query species. For instance, using a human-chimp- macaque alignment to get the ancestral state of human, the human-chimp ancestor sequence is compared to the chimp and to the human-chimp-macaque ancestor. A high-confidence call is made whn all three sequences agree. If the ancestral sequence agrees with one of the other two sequences only, we tag the call as a low-confidence call. If there is more disagreement, the call is not made.
The convention for the sequence is: ACTG : high-confidence call, ancestral state supproted by the other two sequences actg : low-confindence call, ancestral state supported by one sequence only N : failure, the ancestral state is not supported by any other sequence
- : the extant species contains an insertion at this postion . : no coverage in the alignment
You should find a summary.txt file, which contains statistics about the quality of the calls.
I also noted that the fasta file only mask with uppercase and lowercase（and '-', '.'), my qusetion is If lowercase represents multiple states, then how do I distinguish them (eg, how do I distinguish in lowercase regions which are compared to chimpanzees? Which are compared to macaque monkeys? Like REAME as said in.
