search

bcm-uga / Loter

A software package for local ancestry inference and haplotype phasing
Other
38 stars 7 forks source link

Output haplotype order #2

Closed gaushi closed 5 years ago

gaushi commented 6 years ago

Hi Michael, I needed some clarification on how the output haplotypes are ordered. The manual says: "Finally, we can save the result. Each line correspond to an haplotype. The SNPs are space delimited"

Do the haplotypes follow order(two per individual, for diploid organism) exactly similar to the input vcf file?

Thanks a lot, With best regards, Gautam

mblumuga commented 6 years ago

Yes, haplotypes follow order(two per individual, for diploid organism) exactly similar to the input vcf file. Sorry for the slow reply. Michael

meddly commented 5 years ago

Hello Michael, I obtained the results for one admixed individual of two parental populations. I have some questions regarding how to interpret the output. 1- Each color represents one parental, right? May I know which color corresponds to which parental? (code: res_loter = lc.loter_smooth(l_H=[parent1_hap, parent2_hap], h_adm=amx_hap, num_threads=8) loter_chunks_tryp

2- In the output text. Each par of numbers represent the ancestry of the two allels of each SNP?. I would like to know which is the ancestry of each SNP or of a set of SNPs from my VCF. How can I obtain that data from the txt file? image

Best regards, Meddly

gdurif commented 5 years ago

Hi,

The zero corresponds to the first population (index 0) and 1 to the second population (index 1) from the input list l_H=[parent1_hap, parent2_hap]. If you reverse the order l_H=[parent2_hap, parent1_hap] in the input, the 0 and 1 will be inverted in the output.

So you are visualizing the ancestry chunk for the admixed individual where X axis are SNPs (corresponding to columns in res_loter) and Y axis are admixed individuals (corresponding to rows in res_loter). Each color represent a reference population index. To see the matchig between color and ancestry, you can use

plt.imshow(res_loter, interpolation='nearest', aspect='auto')
plt.colorbar()
plt.show()

In the text results, it is the same, rows correspond to admixed individuals in input (in the same order) and columns corresponds to SNPs (again in the same order). So each element of the matrix in the text version gives you the ancetry (0 for pop1 and 1 for pop2) for the corresponding SNPs in the corresponding admixed individuals.

I hope this helps.

ana090192 commented 4 years ago

Hi, Could someone send me a link to the Loter manual? Is there a PDF version?

Thanks a lot, Best, Ana

gdurif commented 4 years ago

Hi Ana, here is the link to the Python notebook explaining how to use Loter (https://github.com/bcm-uga/Loter/blob/master/python-package/Local_Ancestry_Example.ipynb), there is no pdf version for the moment. Best

ana090192 commented 4 years ago

Hi, Thank you very much! Best