Today I thought I would get into a bit about the bizarre beasts in the alveolate group (Fig. 1). Chances are, the term "alveolates" means nothing to you unless you care about the classification of eukaryotes on a broad scale. However, you will certainly be familiar with some of the organisms that are within this group, which includes ciliates (e.g. Tetrahymena), apicomplexan parasites (e.g. Plasmodium, Toxoplasm), dinoflagellates (responsible for red tides) and a couple of other odds and ends that are not important for today's topic but have features heavily in recent debates over the evolutionary history of the lineages.
Figure 1. The major groups of organisms that make up the Alveolata and their evolutionary relationships to one another. The branch leading to Chromera is dashed, indicating uncertainty of its exact affinity.
What is really interesting about this group of organisms is that the all seem to have something really strange going on with their genomic structure. Ciliates contain two nuclei within their single-celled selves, one of which functions as the germline nucleus (called the micronucleus - MIC) whereas the other is the somatic nucleus (macronucleus - MAC). The MAC is the transcriptionally active nucleus and is a highly-processed version of the "silent" MIC and can range between ~120 chromosomes large chromosomes to over 25 million, gene-sized chromosomes, depending on the ciliate. In all known cases, the MAC of daughter cells is produced in a complicated RNA interference-based process of comparison between the parental MAC and the MIC. The processing of genetic information in ciliates is a fascinating subject that science is only beginning to figure out.
Apicomplexans are another unusual group, but I've already discussed them a bit in the past. The moral of the story is that you shouldn't tease little algal cells because you never know when they'll evolve into parasites and cause millions of human deaths annually.
Finally, you have the dinoflagellates. Even though the Amoeba nucleus is the Godzilla of genomes (~670,000 Mb), some dinoflagellates could make a case that their nuclei are at least the Mothra of genomes. Gonyaulax, for instance, has an estimated nuclear size of 98,000 Mb - a mere 34x the size of the human genome. While this alone puts dinos in an unusual genomic category and pretty much ensures that it will be a while before we see a dino nuclear genome completed, it is their plastid (chloroplast) genomes that are truly messed up. Rather than having a single circular plastid genome containing all of the plastid-encoded genes, dinos have fractured their highly reduced plastid genomes into hundreds of plasmids, called minicircles. Only a few genes are individually encoded in multiple copies on these circles and a large proportion of the minicircles are non-coding, containing only the screwed up results of recombinations between different copies of genes. Each minicircle has a a core region, which presumably functions as the site of transcription. Clearly the system has evolved in response to some genomic pressure, but what the possible advantage is of having hundreds of divergent copies of a few genes and a large amount of junk being transcribed in your plastid is, remains a mystery.