Plant breeders are often reluctant to select purely on genomic predictions in rapid cycling generations. It's refreshing to see @kwsgroup.bsky.social now highlighting this as an important approach to accelerate breeding. There are challenges, but I'm convinced this is coming!

Presented on the frontiers of genomic prediction for #PlantBreeding at the 8th Cereal Biotennology and Breeding conference in Budapest, and got some traction on X!

Good point. Collecting the data often need many years, hours and $$$, it adds another irony to use a quick & dirty method for analysis. Machine learners are proud to spend thousands of hours HPC time for an analysis, biologist are proud if they can do an analysis in 30 secs on their laptop.

No, you will not triple yield by tripling the number of grains per spike 👇

1/ If you're in bioinformatics, you're staring at matrices all day. RNA-seq? Gene x sample. scRNA-seq? Gene x cell. Everything is a matrix. But I never learned how to think in matrices. And I regret it.

Confused about vector resize() and reserve() in #C++? Resize modifies the size and fills default values; after resize'ing use [] to fill elements. Reserve only reserves memory in the background but size remains zero; after reserve use push_back to fill the vector. Don't confuse these two!

Somebody has fenced off and seems to claim chromosome X. Elon? 🤣

And another touch of geniality: he used observable Mendelian traits (colors on the seeds) as markers to map the quantitative trait, and understood it needs a cross to prove genetic linkage. It's gene mapping using biparental families in a rudimentary form.

Teaching on additive, dominance, epistatic, transcriptomic, environmental (and more) relationship matrices, it occurs that the VanRaden GRMs are an exception with a sum(2pq) scaling, whereas all others are simply scaled to average diagonal 1. Can't we just scale all 'kernels' to mean diagonal 1?

Unraveling Mendel’s final mysteries: A genomic retelling of pea genetics: Cell www.cell.com/cell/fulltex...

👇A very needed paper and discussion: genomic and phenomic prediction should not be compared, and suggesting that better phenotype prediction will improve breeding is misleading. Phenotypes include environment, but breeders don't sell environment (let that sink in 🤔🙂).

On the domestication of maize (aka corn) with @jrossibarra.bsky.social (a podcast) 👇

Can recommend both the conference and the city 👇

We know GBLUP prediction accuracy gets worse for more distant generations (implying SNP effects changing!). In my teaching I explain it's long-range LD (from selection and family structure) breaking down. This nice article substantiates this more properly: doi.org/10.1186/s127...

This professor position can fit any candidates working on genomic tools to address resilience, sustainability, climate adaptation or biodiversity in agriculture 👇

Our summer course on #genomicPrediction with me and @guillramstein.bsky.social is running this year 4-8 August, application deadline 18 March. Info and registration see: international.au.dk/education/ad...

Interesting work on combining distantly related populations aka closely related species 👇

From "X": New article out by Marta Malinowska et al., including a small gene mapping using our Bayesian mapping tools - soon to be released (keep watching bsky 😉): Automated seminal root angle measurement with corrective annotation 🌱 : doi.org/10.1093/aobp... #PlantBreeding #Automation #Phenotyping