Wouter van der Bijl

The next generation of colour pattern genomics - Nature Ecology & Evolution Convolutional neural networks and genetic association analysis decode the evolution of colour pattern diversity and its underlying complex genetic architecture in the Trinidadian guppy.

‪A really lovely commentary on @vdbijl.bsky.social's paper by Nidal Karagic and Claudius Kratochwil in @natecoevo.nature.com www.nature.com/articles/s41...

Thanks Will, hope you’re doing well!

Deep learning reveals the complex genetic architecture of male guppy colouration - Nature Ecology & Evolution Guppy colouration is a very diverse trait under natural and sexual selection. Using a combination of high-resolution phenotyping and genomic analysis, the authors reveal the genetic architecture of th...

Guppy colouration is a very diverse trait under natural and sexual selection. Using a combination of high-resolution phenotyping and genomic analysis, the authors reveal the genetic architecture of this complex trait www.nature.com/articles/s41...

This is amazing work and a great thread, revealing the hidden genetic complexity behind the equally complex pigmentation patterns. Check it out! 🧪

Yes I’ll be there.

Thanks a lot Alex!! Guppies continue to be amazing.

Thanks Chris!

Thanks Stephen!! And congrats to you on the new gig!

This paper is the result of years of hard work, and would not have been possible without all the help from my co-authors. Special thanks to @jacejostles.bsky.social for taking ~8k photos (+fig 1a), and to @judithmank.bsky.social for her unwavering support of this project.

Overall, the color patterns are made up of a patchwork of ornaments. The genetic architecture is much more complex than previously appreciated.

Duplications from the autosomes to the sex chromosomes play a large role in the evolution of the extreme variation in this species.

Large-effect CNVs underlie sex-linked variation in colour pattern. a, GWAS peaks on autosomes with significant loci with sex-linked inheritance, near texim and in prex1. b, The relative coverage (1× = genome-wide sample average) across the region for 297 males (this study) and 3 females, calculated in moving windows of 1 kb, 100 bp apart. c, Phenotypic effects of the top variant (diamond in a) on ornaments O6 (orange pattern) and B5 and B7 (black pattern). The proportion of variation in the phenotype that is explained by the top SNP is displayed as Nagelkerke’s R2 from a logistic regression. d, Median estimated allele counts for the top SNP, based on read depth, for homozygous reference and heterozygous individuals. No heterozygous alternate individuals were found. e, PheWAS heat maps displaying z scores for the effect of the top variant on orange and black colour across the body, controlling for autosomal and sex-linked relatedness.

The sex-linked ornaments instead show loci of very large effect at copy number variants. The locus of largest effect for orange pattern is at texim, a gene with many duplications to the Y. For black, the largest effect loci is prex1, an X-linked CNV with a know role in NCC migration.

A Manhattan plot showing the association of genomic variants with ornament O2. Points represent genomic variants, with their genomic location on the x axis and the two-sided P value of the association on the y axis (from mixed effects GWAS). P values are displayed as uncorrected for multiple testing, with the dotted line denoting the 5% FDR level.

As the first figure, but for ornament B1.

Next, we sequenced 300 males to map this variation to the genome.

First, for autosomal ornaments we find a polygenic architecture. Many of the loci are near genes involved in neural crest cell migration. These cells later develop into the chromatophores responsible for coloration.

Orange ornaments are strongly heritable, and their incidence and size responded to selection. a, Pictograms of seven orange ornaments. b, Heritabilities of the incidence and size (when present) of each ornament. Dots and lines reflect point estimates and 95% credible intervals, and the gradient bars show the cumulative posterior density. c,d, The effect of selection on the incidence and size of orange ornaments. Bars and points represent means, with error bars showing 95% bootstrapped confidence intervals. x axes show consecutive generations. Sample size (number of fish); P generation: 300; downselected F 1, F2 and F3: 453, 570 and 441; upselected F 1, F2 and F3: 508, 513 and 444.

As in the first figure.

Among all this variation, we can distinguish 7 orange and 8 black ornaments. Together, these can combine into 32k different patterns, although we observe just 691.

Each of the ornaments have different contributions from the autosomes and the sex chromosomes. Only 2 show clear Y-linked inheritance.

The localized heritability of colour is high, but the architecture is heterogeneous. a,b, The narrow-sense heritability of the presence of orange (a) and black (b) colour on the guppy body. For each heat map cell, a Bayesian animal model was used to estimate the contribution of additive genetic and environmental factors. Heritabilities are expressed as a proportion of the between-individual phenotypic variance. The autosomal, X-linked and Y-linked heritability add up to the total heritability. Body positions where the incidence of colour is less than 1% are coloured grey.

Combining the pedigree with >14k guppy photos, we can model the inheritance of ornamentation across the body. Localized heritability is very high overall. Each body location has its own genetic architecture. We see clear X- and Y-linked inheritance, but only for a minority of the body plan.

| Quantification of pattern variation. a, An overview of the phenotyping pipeline. From each photo, the fish is extracted, the body is aligned to a common reference shape using landmarks, and the position of orange and black colour is extracted.

| Orange colour responds rapidly to selection. a, Density plots show the distribution of the percentage of orange colouration per generation and selection regime. Thin lines between generations connect fathers to their sons. Thick lines inside the densities show the median value of colour area.

We set up a selection experiment and phenotyped 3,200 males in a pedigree. Using a deep learning pipeline, we extracted all the ornamentation and aligned the images to a common reference shape.

In just three generations, our selection lines differed by 2x, confirming high heritability.

More guppy males with unique patterns. Photos by Wouter van der Bijl & Jacelyn Shu.

We see 100s of unique color patterns in just a single guppy stock tank in the lab. It has often been said that these patterns are inherited as a single Y-linked locus. Yet, testosterone-treated females also develop ornaments.

So how does the genome encode this variation? Is it really all on the Y?

Photo of guppy males from lab population, each showing a unique combination of black and orange ornaments. Photo by Wouter van der Bijl.

🚨 Super exited to see our paper on the inheritance and genetic basis of guppy color variation come out in @natecoevo.nature.com‬ rdcu.be/eugWV

Guppy males have enormous variation in color patterns, with many combinations of ornamental spots and stripes. But where does all this variation come from?

Go check out @jacejostles.bsky.social 's very pretty poster this evening about the science behind guppy magic and science communication. Spot I1 #evol2025

The aim of this proposal is to place population genomic insights into a comparative framework to gain fundamental insights into the determinants of evolutionary outcomes. The project will work within LepEU, the European Lepidopteran Population Genomics Consortium (https://lepeu.github.io/). LepEU provides access to field samples from European populations of diverse species. Chromosome-scale reference genomes are provided by Project Psyche (https://www.projectpsyche.org/). Networking during the postdoc will be facilitated by participation in the 10kLepGenomes COST Action (https://10klepgenomes.eu/). Existing datasets await analysis, while additional samples need DNA extraction and submission for sequencing. Functional validation capability (CRISPR/Cas9 gene manipulations) is also available to test emergent hypotheses of allele-to-phenotype impacts. Personal research interests of the postdoc will be important to determine the exact project, as the project has a generous sequencing budget.

The successful applicant should have a PhD (obtained within 6 years of the application deadline) in a suitable subject area, such as evolutionary biology or population genomics. A strong interest in population genomics, local adaptation, comparative analyses, and experience working with genomic-scale data is essential. The candidate must have a documented publication record demonstrating relevant skills. Experience working with bioinformatic pipelines (e.g., Snakemake), or working with butterflies is welcome but not essential. The net salary is 28,000 SEK/month (~2,430 Euro, not subject to Swedish income tax) and comes directly from the Carl-Trygger Foundation stipend, which is paid out directly to the postdoc. Only PhD candidates acquired outside of the host department can apply. Currently, the lab of Prof. Wheat consists of 3 postdoctoral researchers, while the Dept. of Zoology provides a vibrant and excellent research environment of active, dynamic researchers. Applications should include: i) a succinct description of research interests and experience, detailing your contribution to any relevant publications (max 1 page), ii) why you are the ideal candidate for this position in the lab (max 1 page); iii) a CV including a list of publications, and iv) the name and contact information of two personal references. Applications will be reviewed on a rolling basis with a deadline of 23 August 2025. The project is planned to start on 1 October, but flexibility in the starting date can be provided for a suitable candidate. Please contact Prof. Wheat for additional information.

🚨Postdoc opportunity🚨: LepEU postdoc: comparative population genomics of European scale adaptation in butterflies

2 year, full-time PD in my group, Stockholm Univ.

Applications assed on rolling basis, deadline: 23 August 2025. Planned start 1 Oct.

Details:
christopherwheatlab.wordpress.com

I'll be at #Evol2025, and would love to talk if you have submission ideas for the journal!!

Am looking forward to being back in Athens for #Evol2025! Come see me talk about @vdbijl.bsky.social 's beautiful work on beautiful guppy colours on Saturday, 11:15 in the Athens GH. www.biorxiv.org/content/10.1...

🚨Anyone want a job?🚨
We have two #postdocs up for grabs! 🧪
- cell developmental biology/#evodevo/#neuroevodevo
- bioinformatics and molecular biology
Both working on brain evolution in Heliconiini butterflies
Details below! Please repost 🙏 1/n

Reading a paper in ProcB that declares AI use as: “We have used AI-assisted technology in creating this article”.

What’s the point of including such an unspecified declaration? It seems equivalent to declaring “we have competing interests” but not declaring what they are.

Wellicht kunt u wat duidelijker communiceren terwijl u hele politieke partijen wegzet as antisemieten.

GroenLinks pvda heeft de tweestatenoplossing in haar verkiezingsprogramma staan.

Interested in a PhD connecting sensory ecology and evolutionary genetics? Applications are now open for a project on the Speciation Genomics of Eye Size Variation in Heliconius Butterflies in our lab at LMU Munich: www.evol.bio.lmu.de/research/mer... Please repost!

Four images to illustrate some prominent single-gene myths. Top left shows a photograph of a person deftly rolling their tongue into a U-shape. Top right shows a photograph of a person’s ear, highlighting the shape and features of the earlobe and cartilage. Bottom left shows a close-up photograph of a person’s eye, with a vivid blue colouration. Bottom right shows a photograph of a person poised to write with their left hand on the blank white page of a spiral-bound notebook.

Remember when you first learned about genetics at school? All those fascinating examples of human traits that are each apparently determined by just a single gene? Time to check in on some of your favourites to see how they’re doing. 🧬🧵🧪 1/n

Biosciences Computing Manager Staff - Non Union Job Category M&P - AAPS Job Profile AAPS Salaried - Information Systems and Technology, Level D Job Title Biosciences Computing Manager Department Administrative Leadership | Dep...

We are looking for new manager for our computing unit! If you have experience in both biology and computers, this might be the perfect job for you! UBC is a lovely place, and Zoology is a fantastic, collaborative and kind department. Apply by May 4th at ubc.wd10.myworkdayjobs.com/ubcstaffjobs...

There should be no question about this!