How to interpret the all results like as 'AAAA'

[liuji70520]

Dear Paul,

Here is the GPT on AAAA and other results of the explanation, but it should be wrong, the correct interpretation should be how it is

Sample Information
P1: Parent 1, the first parental species involved in the hybrid analysis.
Hybrid: The hybrid offspring, which result from the mating of P1 and P2.
P2: Parent 2, the second parental species in the hybrid analysis.
Statistical Parameters
Zscore: Z-score, used to measure the statistical significance of the hybridization event. A higher Z-score indicates that the genotype distribution for this hybrid combination differs significantly from what would be expected by random chance.
Pvalue: P-value, also used to assess the statistical significance of the hybridization event. Typically, a low p-value (e.g., < 0.05 or close to 0) indicates that gene flow is statistically significant.
Gamma: Gamma value, which measures the intensity or proportion of gene flow between P1 and P2. A higher gamma value indicates a greater level of gene flow or introgression from one of the parental species into the hybrid offspring.
Genotype Distribution (AAAA, AAAB, etc.)
The genotype distribution provides detailed counts of different genotypes found in the hybrid offspring. Each genotype format shows a particular combination of gene fragments, where "A" represents fragments from P1 and "B" represents fragments from P2:

AAAA: All gene fragments come from P1 (i.e., all four alleles are derived from P1).
AAAB: Three alleles are from P1, and one is from P2.
AABA: Two alleles are from P1, with one of these occurring at an additional position; the remaining allele is from P2.
AABB: Two alleles from P1 and two from P2, showing a mixture of genes.
AABC, ABAA, ABAB, ABAC and similar: These represent more complex genotype distributions, possibly involving rare combinations or mutations, each showing unique distributions of P1 and P2 gene segments.
ABBA: This genotype represents a symmetric distribution of fragments from P1 and P2.
BAAA: Indicates that most alleles are from P2, with a single allele from P1.
ABBC, CABC, BACA, BCAA and similar: These even more complex distributions suggest diverse gene combinations, possibly indicating mutational events or rare genotype combinations.
ABCD: All four alleles are different, potentially reflecting multiple hybridization events or genotypic variation.

Best regards,
Liu

[liuji70520]

Thank you. I think I probably know what these patterns mean. I want to confirm whether the corresponding order of these patterns is O-P1-hybrid-P2. If so, can I use ABAB-ABBA here to calculate the D statistic (ABBA-ABAB). In Hyde's result, the order of ABAB is O-P1-hybrid-P2, but when introducing the D statistic, the order should be H1-H2-H3-O, right? So the order is reversed. If you can answer me, I will be very grateful to you.

[pblischak]

Hi Liu,

The AAAA, AAAB, etc., columns in the output are the cumulative count of sites with the 16 different possible combinations of nucleotide patterns -- the ChatGPT response is pretty inaccurate. Each row of the output is for a different (P1, Hyb, P2) triplet. You can certainly use these counts to calculate the ABBA-BABA statistic, and there is actually a function to do that in the HyDeResult class (HydeResult.abba_baba)

Let me know if you have any other questions

Thanks,
Paul

[liuji70520]

Dear Paul,

First, thank you for your previous answer.

In my research, I have 100 species, and the map I assigned for each species is its own name, such as "speciesA speciesA." I aim to investigate the distribution of gamma values between genera (P1 and Hybrid belonging to different genera) and within the same genus.

Question 1: Do I need to modify the map file or can I add genus information during subsequent data processing?

Question 2: In the study of introgression proportions within the same genus (or between different genera), should I consider species that were not judged to have significant hybridization events (since I only used hyde-out-filtered.txt)? For example, if the average gamma value for genus A (within the same genus) is 0.25, I interpret this as 25% or more of the species within this genus having introgression from the same genus. However, if we consider species that were not deemed to have significant hybridization events, this value would likely decrease significantly.

Here, I have this idea because you mentioned in the text, 'if we are testing four diploid individuals in a putative hybrid and only two of them are 50:50 hybrids ( = 0.5 for each), then the value of gamma for the whole population will be = 0.25.' My understanding is that if we treat the entire genus as one population, when calculating the average gamma value between different genera, do we need to divide by the total number of species in the genus?

Thank you again for your guidance!

Best regards,
Liu