gregorgorjanc.bsky.social
Managing and improving populations using data science, genetics and breeding: @HighlanderLab, @RoslinInstitute, & @TheDickVet.
https://www.ed.ac.uk/roslin/highlanderlab
157 posts
307 followers
336 following
Regular Contributor
Active Commenter
comment in response to
post
Lastly, I charied the session on Selection and Response, with an introduction to the topic and leading discussions for the talks by Christopher Wheat and Michael Morrissey.
comment in response to
post
Guilherme da Silva Pereira presented a poster led by Nathállia Pires Nogueira: Using tree sequences to study genetic relationships among individuals in a tetraploid potato breeding population
comment in response to
post
Maybe you run some simulations where you know the true genetic and environmental correlations and convince yourself what is happening. There is a couple of great QG simulators out there. Since you are in R, I can recommend our AlphaSimR;)
comment in response to
post
What kind of scaling do you use? Quan Gen parameters can be impacted by scaling, particularly those connected to interactions, though you are not doing that, I think.
comment in response to
post
Lastly I gave a short course on Stochastic simulations of breeding programmes with AlphaSimR - a teaser for our free on-line course www.edx.org/learn/animal...
comment in response to
post
I presented the work of Letícia Lara on Evaluation of selective breeding programme designs for black soldier fly larvae body weight (with María Martínez Castillero, Thiago Oliveira, Ivan Pocrnić, Jana Obšteter)
comment in response to
post
Jana Obšteter presented the work of Laura Strachan on Optimizing pedigree reconstruction and patriline determination in honeybees (with Jernej Bubnič and Janez Presern)
comment in response to
post
Irene de Carlos Fernández presented her work on Modelling the Impact of Non-Native Honey Bee Importation on Native Apis mellifera mellifera Populations (with Laura Strachan, Grace McCormack, Jana Obšteter)
comment in response to
post
It’s a shame this can’t work better I think. Every once in a while I find that some syncing breaks or lags horribly. It should not be that hard to achieve this in the 21st century!
comment in response to
post
It’s a pain. I now forward everything to MS account to have everything in one place.
comment in response to
post
To learn biology from human genetics, we usually want to know the relationship between gene function and phenotype. For example, does a decrease in gene expression increase or decrease the trait?
GWAS doesn't give us this directly, but Loss-of-Function (LoF) tests do.
comment in response to
post
Finally, I've written an #rstats package named Reacnorm, which fully implements our framework in practice, with a (hopefully) simple tutorial using #brms as a vignette for the package.
github.com/devillemereu...
The package is currently on its way to #CRAN.
comment in response to
post
We define different variance components of a plastic trait along some environmental variable : the variance due to the average shape of the reaction norm (V_Plas), the genetic variance (V_Gen) and a residual variance (V_Res).
comment in response to
post
Missed out this time? Join our next instance in May! genomics.ed.ac.uk/event/introd...
comment in response to
post
Some neat examples are pmc.ncbi.nlm.nih.gov/articles/PMC..., pubmed.ncbi.nlm.nih.gov/35201891/, elifesciences.org/articles/72177, academic.oup.com/mbe/article/..., …
comment in response to
post
In cross-validation for yield prediction, the ARG-based branch relationship matrix (BRM) demonstrated higher predictive ability than the standard site-based relationship matrix (SRM) when combining both subspecies.
comment in response to
post
GWAS hits with SRM (A) and BRM (B) were similar
comment in response to
post
(A) The standard site-based relationship matrix (SRM, VanRaden’s) and (B) the ARG-based branch relationship matrix (BRM) revealed similar population structure, with highly correlated (C) diagonal and (D) off-diagonal elements, though on different scales.
comment in response to
post
The age distributions for (A) nodes (ancestors), (B) mutations, and (C) SNP sites (i.e., first mutation at each site) were heavily right-skewed towards the present (as expected).
comment in response to
post
The ARG encoded genomic data more efficiently than the standard VCF: the tree sequence file for all chromosomes was 62 MB, compared to 228 MB for the VCF—nearly four times smaller!
comment in response to
post
Local trees from two genomic regions showed distinct patterns: (A) revealed deep separation between indica and japonica, linked to the DST gene associated with panicle length in japonica. The (B) region segregated in both subspecies and was linked to panicle traits in both.
comment in response to
post
After building the ARG, we demonstrated it captures biological signals using genealogical nearest neighbors (GNN) - it clearly distinguished indica and japonica rice subspecies and effectively represented population structure.
comment in response to
post
This builds upon the work of tskit community - there is an advanced manuscript in the pipeline on how to build relationship/relatedness matrix from tree sequence encoding of an ARG!
comment in response to
post
Taken together, the results demonstrated the potential of ARGs for the quantitative genetic analysis of diverse populations.
comment in response to
post
The ARG:
* Captured key biological signals
* Encoded genomic data more efficiently than conventional formats
* Resulted in the highest predictive ability when combining both subspecies (though not by much)
comment in response to
post
We evaluated the application of an Ancestral Recombination Graph (ARG), encoded as a tree sequence, to characterise a breeding population, and for genomic prediction and genome-wide association studies.
comment in response to
post
This is from the MSc internship of Ines!
comment in response to
post
Xinger (Evie) Tang presented a poster “A Pedigree-Based Method for Localization of Recombination Events”
comment in response to
post
Jaime Ortiz Cuadros presented a poster “Accounting for uncertainty in Optimal Contribution Selection” - joint work with Josh Fogg, Julian Hall, and Ivan Pocrnic
comment in response to
post
Yu (Uni) Zhang presented a poster “How well must we characterise empirical populations for breeding simulations?”
comment in response to
post
Hannes Becher presented a poster “Fast pedigree PCA with the randPedPCA R package”- joint work with @epigenci.bsky.social
comment in response to
post
Ros Craddock talked about “General Pedigree Tracking of Disease-causing Alleles for Recessive Monogenic Diseases in Dogs” - joint work with Joanna ilska, Cathryn Mellersh, Pam Wiener, and Audrey Martin
comment in response to
post
Gabriela Mafra Fortuna talked abou “Experience with Inferring Ancestral Recombination Graph of the Global Cattle Population” - joint work with Jana Obsteter
comment in response to
post
Let us know if you match the essential criteria and are interested in the post.