Yath

The ocean is one of the last great frontiers of biology. More than 90% of all marine life (an estimated 2M species) have not been discovered. Also, only ~10% of the seafloor has been mapped at 100m resolution, "as compared to 100% of...Mars and Venus." We should sequence everything in the ocean. Doing so would lead to many useful biotech tools, just as the thermostable polymerase used for PCR was found in a geyser-dwelling microbe, or CRISPR was identified in microbes living in Mediterranean salt marshes. With this in mind, I've been reading a lot more about John O. Dabiri's work at Caltech. His report on "Bio-inspired Ocean Exploration," in particular, is a must-read. It has so many good vignettes about how little of the ocean we understand, and why it's so difficult to study. "The ocean represents...99% of the habitable volume on Earth," the report says, "and yet, less than one-tenth of one percent of the seafloor has been mapped" at a resolution of 1 meter. The obvious plan to sequence the ocean, of course, would be to make a bunch of submarines and underwater vehicles and send them out with nanopore sequencers. But this would not viably scale. For example: "...it has been estimated that it would require 200 shipyears to sample the entire ocean at just one depth, that is, a single ship moving through the ocean for 200 years, or a fleet of 200 ships, all working in concert for a whole year—just to measure one depth in the ocean.” A typical, propeller-driven underwater vehicle costs about $50,000. And here's the bit that really astounds me: "...the volume of ocean water is roughly one billion cubic kilometers. A uniformly distributed array of one million measurement systems would therefore each still be tasked with monitoring an area equal to the size of the city of Los Angeles in the United States (1,000 km2) and throughout a depth of one kilometer of the ocean. By this thought exercise, it becomes apparent that even one million sensors would likely be a significant underestimate for the task at hand.” Dabiri's group is attaching tiny microcontrollers and batteries to living jellyfish, and then using the animals to explore the ocean. Jellyfish don't have a swim bladder, so they "can be found as deep as the Marianas Trench, at least 3,700 m below the ocean surface.” Each jellyfish is steered through the water using pulses of electricity. Jellyfish don’t have a central nervous system or pain receptors, so they probably don’t suffer. In the lab, they didn’t show any signs of stress or impaired feeding, reproduction, etc. The microcontroller and battery are placed in the center of its bell, with electrodes wired out toward its extremities. Each unit costs a few dollars, and takes "less than a few seconds per animal" to implant. Not saying this is necessarily ethical, but it is a remarkably clever way to improve the cost & coverage of ocean mapping (and maybe, soon, sequencing?) by orders of magnitude.