Giant kelp (Macrocystis pyrifera) is a prominent marine organism that plays a significant role in maintaining the ecological balance of coastal ecosystems (Graham et al., 2007; Reed and Brzezinski, 2009). As the largest and fastest-growing species of algae on Earth, it provides habitat, food, and shelter for a diverse range of marine species (Edwards et al., 2015). However, giant kelp populations are facing unprecedented challenges due to increasingly frequent and severe heatwaves. These heatwaves have negatively impacted giant kelp, leading to episodes of mass die-offs and threatening its genetic diversity (Filbee-Dexter et al., 2020; Smale, 2020). Beyond its ecological significance, giant kelp holds immense promise for human benefit, particularly in the fields of kelp farming and ecosystem restoration (Campbell et al., 2014; Buschmann et al., 2017; Layton et al., 2020).

Giant kelp farming has garnered growing interest as an alternative means of bolstering the world’s food supply, producing biofuels, and sequestering carbon dioxide from the atmosphere (Duarte et al., 2021; Spillias et al., 2023). The cultivation of giant kelp has demonstrated remarkable potential for sustainable aquaculture practices due to its rapid growth rates, ability to absorb excess nutrients from surrounding waters, and its lack of competition for agricultural land, fertilizer, or freshwater resources (Gerard, 1982; Stewart et al., 2009; Rassweiler et al., 2018). Recent advances in kelp genetics, including the release of many high quality annotated genomes, are helping to unlock the vast potential of genomics in improving giant kelp commercial cultivation and restoration efforts (Paul et al., 2022; Denoeud et al., 2024).

Genetic resources, including high-quality annotated reference genomes and a sequenced founding populations, are essential for advancing breeding programs and maximizing desirable traits in crops (Galluzzi et al., 2020; Pour-Aboughadareh et al., 2021). By utilizing comprehensive genetic data, scientists and farmers can identify and select individual kelp specimens with favorable traits, such as enhanced growth rates, improved nutritional content, and heightened resistance to environmental stressors (Briggs, 1998). Genetic information could aid in developing novel techniques to cultivate kelp strains better adapted to diverse oceanic conditions and various aquaculture systems. Moreover, studying naturally occurring mutations provides an alternative to artificially produced mutations and offers a valuable approach to addressing some concerns associated with genetically modified organisms, with benefits in permitting, safety perception, and preserving genetic diversity. Over 60 newly annotated brown macroalgae reference genomes, including two giant kelp genomes, greatly increase the genetic resources available to aid in giant kelp breeding. However, the lack of whole-genome mutational data from many individuals across different giant kelp populations hinders genomics-assisted breeding efforts (Paul et al., 2022; Diesel et al., 2023; Denoeud et al., 2024).

This research contributes to genomic tools bridging the macroalgae knowledge gap by providing a detailed study of the mutation landscape of giant kelp. It aims to uncover insights into mutation patterns, selective pressures, and create a mutation effect database for a seedbank, thereby facilitating breeding efforts and scientific research. By understanding the genetic basis of desirable traits and naturally occurring mutations, this research aims to pave the way for the sustainable and efficient cultivation of giant kelp while contributing to the preservation of marine biodiversity in the face of environmental challenges.

We classified single nucleotide polymorphisms (SNPs) from 559 individuals from four populations (Catalina Island [CI], Camp Pendleton [CP], Arroyo Quemado [AQ], and Leo Carrillo [LC]) into modifier, low-impact, moderate-impact, and high-impact mutations based on their predicted effects on gene sequences, according to SnpEff list of effects (Cingolani et al., 2012). All four populations are located within a narrow band between 11 m and 1.8 km from the shoreline of Southern California, with yearly average sea surface temperatures ranging from 15.5°C to 17.7°C ( Table 1 ), and have previously been identified as distinct populations (Johansson et al., 2015). We found over 28 million variants, resulting in more than 42 million predicted effects. Of those, 97% were modifiers, expected to have no significant impact on a gene. Low- and moderate-impact mutations account for 1.5% and 1.4% of mutations, respectively, while high-impact mutations make up 0.05% of all mutations ( Table 2 ). The average number of low-, moderate-, and high-impact mutations per individual is approximately 36,100, 33,300, and 6,000, respectively ( Figure 1 ). The Catalina Island population exhibited a disproportionately higher ratio of high-impact mutations compared to the other populations ( Figure 2 ). Further investigation into the underlying causes of this disparity could provide valuable insights into the genetic adaptations of giant kelp in response to local conditions.

Genome annotation often relies on comparative genomics, where similarities between the target species’ genes and those of well-annotated model species are used to infer gene function. In the case of brown macroalgae like giant kelp, recent advances in kelp genomics increase the resources available for comparative studies (Cormier et al., 2017; Shan et al., 2020; Paul et al., 2022; Denoeud et al., 2024). However, the novel pathways found in brown algae, such as alginate metabolism, require additional validation (Mazéas et al., 2024). While increasing sequencing data for kelp is important, integrating experimental approaches—such as gene knockdowns, overexpression studies, or associating specific genes with phenotypic traits—will be key to advancing our understanding of kelp’s gene functions. The giant kelp genome contains 25,900 genes, but only about 8,000 have known annotated functions and are mapped to Eukaryotic Orthologous Groups (KOG). Here, annotated genes tend to be larger in size and have fewer impactful mutations compared to genes that are not annotated ( Figure 3 ). This suggests a higher degree of conservation among annotated genes, likely due to factors such as ascertainment bias in annotation and the unique functional specialization of giant kelp, which diverges from traditional model organisms (Armengaud et al., 2014; Parikesit et al., 2014).

Giant kelp holds great ecological and economic significance, serving as a cornerstone species in marine ecosystems and supporting diverse marine life. However, it faces challenges including habitat degradation, climate change impacts, and a lack of the genetic resources required for effective restoration efforts and to maximize its farming potential. To help address this issue, we established a comprehensive database of naturally occurring mutations in a seedbank containing 559 individuals and explored the effects of these gene mutations. We annotated over 15.9 million mutations in 25,900 genes of the giant kelp genome. While the results can provide valuable insights, it is important to note the limitations of this dataset, as only 8,000 of the genes have functional annotation and are placed in a KOG pathway. This underscores the need for improved gene annotation methodologies for brown macroalgae like giant kelp. Recent genomic advances in brown macroalgae research have the potential to significantly improve predictions of mutation effects and enhance the use of natural knockouts.

In conclusion, establishing this comprehensive mutation database represents a significant milestone in conserving and sustainably managing giant kelp populations. By leveraging genetic insights, we now have a valuable resource to guide restoration efforts and identify preexisting gene knockouts for breeding programs aimed at cultivating resilient and productive kelp farms.

Details on sporophyte collection, gametophyte isolation, DNA extraction and sequencing, farm design, and phenotyping are described in Osborne et al (Osborne et al., 2023). Briefly, sequencing was performed on an Illumina S4 Novaseq platform (150 bp paired-end) at the BGI North American NGS lab, generating approximately 11.2 GB or 87 million reads per sample. Raw reads were then trimmed using fastp (version 0.20.1) (Chen et al., 2018). Trimmed reads were then aligned to the giant kelp reference genome using hisat2 v2.1 with standard parameters (Tajima, 1989). The genome-wide coverage per sample was approximately two to three times the expected depth (Kim et al., 2019; Diesel et al., 2023). Bam files had their duplicates marked using the GATK4 v4.1.2 command “MarkDuplicates”, and then multiple bam files for a single individual genotype were collapsed into a single bam file using Samtools (Li et al., 2009; Van der Auwera et al., 2013).

Genetic variants were called using the GATK4 v4.1.2 with a ploidy set to 1 (Van der Auwera et al., 2013). Individual GVCF files were then merged and converted into a raw VCF file containing variant information and used for downstream applications using GATK v4.1.2 (Van der Auwera et al., 2013). The raw nuclear VCF file containing 559 individuals, was then filtered for downstream genotype-phenotype modeling applications based on vcftools and GATK4 best practices (Van der Auwera et al., 2013; Danecek et al., 2021). Mutation classification was done with Snpeff (Cingolani et al., 2012). A nonstandard database was created using the giant kelps genome, and the General Feature Format (GFF) is publicly available from the JGI algal genome portal PhycoCosm (https://phycocosm.jgi.doe.gov/Macpyr2) (Grigoriev et al., 2021; Diesel et al., 2023). SnpEff was run with standard settings. Bcftools view -r was used to generate gene-specific vcf files (Danecek et al., 2021). For each gene, a vcf file was read into R using vcfR (Knaus and Grünwald, 2017), and hierfstat was used to calculate Fst, nucleotide diversity, and Tajima’s D (Goudet, 2005).