-
Notifications
You must be signed in to change notification settings - Fork 12
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Improve the algorithm of the main peeling function #169
Comments
Have Done @XingerTang :
The "e" defined in line 19 might avioid division by zero in calculations But the question is :If we just hope to avoid zero, why we need to multiply by 1 - e? |
@XingerTang @gregorgorjanc I also summarised the correspondence between the formula in the paper and the objects or variables in the code. Hope it will help modify the code. Formulagenotype
|
@AprilYUZhang @gregorgorjanc In the algorithm, the posterior probabilities are updated with the following variables:
The shape of The shape of The shape of The calculation of the posterior probabilities is started with the calculation of the The reason why the anterior probabilities can be updated inside the main peeling function which is executed family by family while posterior probabilities can only be updated outside the main peeling function and be updated generation by generation is that, each individual can only have one sire or dam however each sire or each dam can have multiple families. To gather the information upward requires calculations across the families in a generation. |
@AprilYUZhang @gregorgorjanc The detailed version of the accuracy:
The full reports:
|
@AprilYUZhang @gregorgorjanc The detailed version of the accuracy:
The full reports:
|
Thank you @AprilYUZhang for the summary! But there are still a few errors in |
@AprilYUZhang @gregorgorjanc The old The new |
I tried to organize the errors of the peeling in addition to the Uni's work above. The sources of the errors during the peeling could be:
For the
Note that the influences of the errors should decrease during the iteration as the information propagates through the pedigree, maybe we can also introduce a mechanism that reduces the size of the errors during the peeling process. |
mutation rate will be an argument for AlphaPeel with a default value set. |
@gregorgorjanc @AprilYUZhang
The following are the steps we can take for the algorithm improvements:
The text was updated successfully, but these errors were encountered: