Unlocking Salmonberry Color: Pool-seq & Fst Gene Discovery

Hey everyone! Ever looked at a juicy salmonberry and wondered what makes it red or gold? Well, you're not alone! We're on an exciting journey, using some cutting-edge genomic tools to crack the genetic code behind the vibrant color polymorphism in Rubus spectabilis, our beloved salmonberry. This isn't just about pretty berries; it's about understanding fundamental genetic control, and it has huge implications for ecology, evolution, and even horticulture. Our quest to pinpoint the specific genes involved has led us deep into the world of Pool-seq and Fst calculations, and while it's been a ride with its fair share of twists and turns, we're here to share our adventure, insights, and some crucial tips for navigating similar genomic expeditions. We're talking about finding those tiny genetic differences—the SNPs with unusually high or low Fst—that could be the key to unlocking the salmonberry's color secrets. So, buckle up, because we're diving into the nitty-gritty of genetic differentiation, troubleshooting parameters, and making sense of some seriously fascinating data. This isn't just a technical discussion; it's a peek into how scientists tackle real-world biological puzzles using powerful computational genomics.

The Mystery of Salmonberry Color: Our Genetic Quest

Alright, guys, let's talk about Rubus spectabilis, affectionately known as salmonberry. It's this incredible native raspberry-relative found throughout the Pacific Northwest, and what makes it super cool, beyond its delicious fruit, is its striking color polymorphism: you can find individuals producing bright red berries and others flaunting gorgeous gold ones. For us, this isn't just a pretty sight; it’s a fascinating genetic puzzle waiting to be solved. Our main hypothesis here is that the gold morphs might be lacking the function of a gene (or genes!) responsible for producing anthocyanins, which are the pigments that give red berries their vibrant hue. Interestingly, we've observed that the fruit color doesn't seem to correlate with specific environmental factors, suggesting a relatively simple genetic control mechanism, which is great news for our research!

To unravel this mystery, we've gone all out, building a haplotype-phased reference genome for both red and gold morph salmonberry. We used state-of-the-art PacBio HiFi and Hi-C data from the Canada BioGenome Project, which is a total game-changer for genome assembly. After building these awesome genomes, we made sure to compare them at a foundational level, checking for things like nucleotide divergence, alignment patterns, and any major structural variations. This crucial step helped us confirm that, despite their color differences, the two morphs aren't entirely distinct 'species' at the genomic level, which validates our approach. For now, we're using the red morph salmonberry genome as our primary reference because our working assumption is that the red phenotype represents the 'wild-type' or functional state for pigment production. This detailed genomic groundwork is absolutely essential before we even start looking for those elusive color-determining genes, setting us up for success in our subsequent analyses, especially when we start calculating Fst to identify regions of high genetic differentiation between our red and gold populations. This deep dive into the foundational genomics ensures that any subsequent SNPs with unusually high or low Fst we identify are truly meaningful signals of differentiation rather than artifacts of poor reference quality or extreme genomic divergence.

Diving Deep with Pool-seq: Our Approach to Finding Color Genes

So, how do you find those specific genetic loci responsible for color? Well, we decided to use a really smart and cost-effective strategy called Pool-seq. For those unfamiliar, Pool-seq involves pooling DNA from many individuals of a particular group (like our red-berried salmonberries) and then sequencing the entire pool. This gives you average allele frequencies for that group, which is super powerful for detecting population-level differences without sequencing each individual separately. It's like taking a snapshot of the genetic makeup of an entire population without the hefty price tag of individual sequencing. We meticulously sampled a bunch of both red and gold individuals, then pooled them strategically by population and color. Our goal was whole-genome sequencing (WGS) of these pools, targeting a solid 4X coverage per individual using Illumina NovaSeq X PE150bp reads. This high coverage is essential to accurately estimate allele frequencies within each pool, especially when dealing with varying pool sizes.

Our experimental design is quite robust, spanning eight populations in total – that's four geographically distinct locations, with paired red versus gold berry pools from each. This geographical replication is key, guys, because it helps us identify genetic signals that are truly consistent across different environments, making our findings much more reliable. However, as is often the case in field sampling, the number of individuals per pool varied. Two of our locations boasted impressive pool sizes of around 100 individuals each for both red and gold morphs, which is fantastic for robust allele frequency estimation. But, for the other two locations, we had smaller pools, with only about 20 individuals each, and the distribution wasn't always equal between red and gold pools. This variation in pool size is an important factor to keep in mind when interpreting our results, as smaller pools can sometimes lead to noisier allele frequency estimates and potentially impact Fst calculations. Despite these practical sampling limitations, the Pool-seq approach remains incredibly valuable for pinpointing regions of the genome that show significant genetic differentiation between red and gold morphs, which we quantify using Fst. Our initial analyses with PoolParty2 showed some promising, albeit inconsistent, peaks. These peaks, representing SNPs with unusually high Fst, hinted at potential loci but lacked the consistency across populations that we were really looking for. That's why we turned to grenedalf, seeking a more robust confirmation and a deeper understanding of our Fst landscape.

The Initial Journey: Exploring PoolParty2's Peaks and Puzzles

When we first embarked on this genetic adventure, we naturally gravitated towards an automated pipeline that seemed perfect for the job: PoolParty2, developed by Micheletti SJ & Narum SR in 2018. This tool is pretty neat and offers a streamlined way to analyze Pool-seq data. We used it to calculate Fst between our red and gold pools, both on a per-location basis and across all four locations combined. The initial results with PoolParty2 were a mixed bag, to be honest. We definitely saw some promising Fst peaks, those exciting SNPs with unusually high Fst that scream