Have we got the power?

Sabrina Heiser wrote this post as a final project for Stacy Krueger-Hadfield’s Science Communication course at the University of Alabama at Birmingham. Sabrina grew up in Germany, completed a BSc (Hons) in Marine Biology at Plymouth University (UK) and then lived in Antarctica for 2.5 years working for the British Antarctic Survey. Now, as a PhD student in Dr. Chuck Amsler’s lab at UAB, she is finally able to combine her love for macroalgae and the Frozen Continent, where she is investigating algal population structure and how gene flow shapes the distribution of geographic patterns in physiological traits. Sabrina tweets at @sabrinaheiser.

Genetic diversity is something we all worry about — especially in our rapidly changing climate. We care about species when trying to conserve biodiversity, but that inherently means we also need to care about the historical and contemporary processes that have resulted in the patterns of genetic diversity within that species. Hoban (2014) predicted that simulation software would be increasingly used to address these issues, but is it?

Simulations can help us calculate and maximize the statistical power of a given sampling strategy. We can optimize power by balancing the amount of populations, genetic markers and individuals that are being sampled (Hoban, 2014). Increasing the number of markers used, can drastically decrease the amount of individuals required, for example. This is especially important when studying organisms with limited access due to location or abundance.

But, what about more complicated organisms, like haploid-diploid species?

In ferns, seaweeds and some fungal taxa, haploids and diploids lead separate, but connected lives. The haploid and diploid stages are morphologically and physiologically distinct, however, they are connected through gene flow through the processes of meiosis and fertilization (see here). What sampling problems arise if haploids and diploids do not live together, such as in ferns (Nitta et al., 2017) or some seaweeds (Krueger-Hadfield et al., 2016)?

Krueger-Hadfield and Hoban (2016) provided some guidelines to incorporate simulations and power analyses into studies that address haploid-diploid population genetics. If the two ploidies are easily distinguishable (i.e., morphologically), keeping track of each ploidy from collection to data analyses is relatively straightforward. However, it becomes a bit more interesting if the life stages are isomorphic with few readily identifiable differences in the field, or even the lab, to distinguish the stages. Unless reproductive structures are present, there is no easy way of telling them apart. So, how do you make sure you are not making haploids into diploids in downstream analyses as most programs and pipelines assume diploidy?

In some species, ploidy levels can be determined using assays in the laboratory, such as chemical tests (Krueger-Hadfield et al., 2011). Others have developed sex/ploidy-linked markers (Guillemin et al., 2012) or used microsatellites to construct multilocus genotypes (Guillemin et al., 2008; Krueger-Hadfield et al., 2013; 2016). If at least one microsatellite locus is heterozygous, then the sample should be a diploid. Fixed homozygosity means it’s likely a haploid. However, there might be rare cases in which all loci are homozygous, but the sampled individual is actually diploid. Hence, having several, independent ways of confirming ploidy is not only advantageous, but critical to understand natural history and population dynamics.

So, how do you go about sampling haploid-diploid populations? One, interesting pattern to study is departures from expected ratios of females to males and haploids to diploids. Males and females are expected to occur at a ratio of 1:1, if meiosis is acting properly in the diploid stage (Destombe et al., 1989). Intuitively, one might think that the same is true for haploid-diploid ratios. However, many species with isomorphic life cycles are also dioicious, in which there are independent male and female individuals (see, for example, Krueger-Hadfield et al., 2015). Therefore, only females are producing offspring and the diploids pay the inherent cost of producing males. If there are no fitness differences, then we should find haploid-diploid ratios of √2:1 (Destombe et al., 1989; Thornber & Gaines, 2004). The closer the haploid-diploid ratio is to √2:1, the larger the sample size needs to be in order to detect deviations (Krueger-Hadfield & Hoban, 2016).

The haploid-diploid ratio, therefore, impacts how you sample to look at gene flow. In order to determine patterns of gene flow between haploids and diploids, it will likely be necessary to sample double the amount of haploids compared to diploids (Krueger-Hadfield & Hoban, 2016). It feels a little bit like the chicken and the egg. How many of each ploidy stage are we collecting? We might not know without some sort of genetic marker. Yet, if we don’t know how many we are collecting of each ploidy, we might not collect the amount of individuals we need and may not have sufficient power. So what comes first? Knowing the ploidy or collecting the right amount of samples?

This is the problem that I have been facing trying to plan my upcoming field season in Antarctica. The samples that I will be collecting are not going to be analysed until I get back to the lab at the University of Alabama at Birmingham. Therefore, I may not know until it’s too late whether I’ve sampled enough haploids and diploids to describe the population genetic structure of an important understory algal species. This is a common problem in studying haploid-diploid species, so I won’t be the only one. For example, Krueger-Hadfield et al. (2016) had only 30 samples from native, Japanese populations in the red seaweed Gracilaria vermiculophylla and were consequently limited in their understanding of the mating system and haploid-diploir ratio variation observed across populations.

Herein lies the problem with studying haploid-diploid species. We need to know about their demography to effectively sample, but there is so little known about population structure and dynamics. Nevertheless, seaweed forests are some of the most productive ecosystems on earth, so we have an urgent need to understand these systems. In the Antarctic, kelp forests match their temperate cousins in biomass (Wiencke & Amsler, 2012), but we know next to nothing about gene flow patterns among the dominant algal species at high latitudes in the Southern Hemisphere. How can we safeguard these habitats and implement policies about marine protected areas without knowing how genetic diversity is generated and maintained?

So, it’s critical we keep power in mind. Overly excessive sampling may do little to increase power. Too little sampling may be too little, too late. If we keep the balance and plan properly, we may just have the power, even in organisms with complex life cycles!


Destombe C., Valero M., Vernet P. & Couvet D. (1989) What controls haploid-diploid ratio in the red alga, Gracilaria verrucosa? Journal of Evolutionary Biology. 2: 317-38.

Guillemin M.-L., Faugeron S., Destombe C., Viard F., Correa J.A. & Valero M. (2008) Genetic variation in wild and cultivated populations of the haploid-diploid red alga Gracilaria chilensis: how farming practices favour asexual reproduction and heterozygosity. Evolution. 62(6): 1500-19.

Guillemin M.-L., Huanel O.R. & Martínez E.A. (2012) Characterization of genetic markers linked to sex determination in the haploid-diploid red alga Gracilaria chilensis. Journal of Phycology. 48(2): 365-72.

Hoban S.M. (2014) An overview of the utility of population simulation software in molecular ecology. Molecular Ecology. 23: 2383-401.

Krueger-Hadfield S.A., Collén J., Daguin-Thiébaut C. & Valero M. (2011) Genetic population structure and mating system in Chondrus crispus (Rhodophyta). Journal of Phycology. 47:440–50.

Krueger-Hadfield S.A., Roze D., Mauger S. & Valero M. (2013) Intergametophytic selfing and microgeographic genetic structure shape populations of the intertidal red seaweed Chondrus crispus. Molecular Ecology. 22:3242–60.

Krueger-Hadfield S.A., Roze D., Correa J.A., Destombe C. & Valero M. (2015) O father where art thou? Paternity analyses in a natural population of the haploid-diploid seaweed Chondrus crispus. Heredity. 114: 185-94.

Krueger-Hadfield S.A. and Hoban S.M. (2016) The importance of effective sampling for exploring the population dynamics of haploid-diploid seaweeds. Journal of Phycology. 52:1-9.

Krueger-Hadfield S.A., Kollars N.M., Byers J.E., Greig T.W., Hamman M., Murray D., Murren C.J., Strand A.E., Terada R., Weinberger F. and Sotka E.E. (2016) Invasion of novel habitats uncouples haplo-diplontic life cycles. Molecular Ecology. 25: 3801-3816.

Nitta J.H., Meyer J.-Y., Taputuarai R. & Davis C.C. (2016) Life cycle matters: DNA barcoding reveals contrasting community structure between fern sporophytes and gametophytes. Ecological Monographs. Online in advance of print.

Thornber C.S. & Gaines S.D. (2004) Population demographics in species with biphasic life cycles. Ecology. 85(6): 1661-74.

Wiencke C. & Amsler C.D. (2012) Seaweeds and their communities in polar regions. In Wiencke C. & Bischof K. [Eds.] Seaweed Biology: Novel Insights into Ecophysiology, Ecology and Utilization. Springer-Verlag, Berlin, pp. 265-94.

About Stacy Krueger-Hadfield

I am a marine evolutionary ecologist interested in the impacts of seascapes and complex life cycles on marine population dynamics. I use natural history, manipulative field experiments and population genetic and genomic approaches with algal and invertebrate models in temperate rocky shores,estuaries and the open ocean.
This entry was posted in blogging, community ecology, evolution, haploid-diploid, natural history, NSF, population genetics and tagged , , , , , , , . Bookmark the permalink.