Bringing Extinct Species Back to Life | Dr. Beth Shapiro

What It Covers

Dr. Beth Shapiro, Chief Scientific Officer at Colossal Biosciences, explains the science behind de-extinction projects targeting woolly mammoths, dodo birds, dire wolves, and thylacines. The conversation spans ancient DNA recovery, CRISPR-based genome editing, species concepts, Neanderthal hybridization in modern humans, ecosystem restoration logic, and the parallel application of these tools to prevent living species from going extinct.

Key Questions Answered

• •De-extinction methodology: Colossal's approach does not recreate an identical genome. Scientists sequence multiple fossil genomes, identify genetic variants shared across specimens but absent in living relatives, then engineer only those specific changes into the closest living relative's genome. For dire wolves, 20 targeted edits to gray wolf DNA produced larger body size, increased musculature, and light-colored coat — traits confirmed from fossil direwolf genomes, not guesswork.
• •Ancient DNA preservation limits: DNA degrades through three processes — UV radiation, freeze-thaw water expansion, and microbial decay. The oldest recoverable DNA comes from permafrost-preserved mammoth bones at roughly 1–2 million years old. Dinosaurs, extinct 66 million years ago, left only fossilized rock with zero recoverable genetic material, making any dinosaur de-extinction biologically impossible regardless of technology advancement.
• •Neanderthal DNA in living humans: Modern humans outside Africa carry 2–5% Neanderthal DNA, but each person carries a different 2–5%. Collectively, living humans preserve over 90–95% of the total Neanderthal genome. The genomic regions where no living person carries Neanderthal variants — roughly 5% — represent genes where the human version was essential for survival, making those regions the primary target for understanding what genetically distinguishes Homo sapiens.
• •Ecosystem redundancy as conservation strategy: Reintroducing apex predators stabilizes ecosystems through trophic cascades. Gray wolf reintroduction in Yellowstone reduced overgrazing, which allowed riverside vegetation to recover and altered river flow patterns. The thylacine's absence from Tasmania correlates with uncontrolled population growth of prey species and the spread of Tasmanian devil facial tumor disease — a transmissible cancer enabled partly by the genetic homogeneity that results from missing predator-driven population pressure.
• •Gene drives for invasive species control: Synthetic biology gene drives — engineered genetic elements that propagate through wild populations across generations — can suppress invasive species without permanent ecosystem removal. Drives can be designed with generation-limited lifespans and built-in off switches, since natural selection strongly favors any organism that breaks the drive and reproduces normally. Shapiro identifies Mediterranean cheatgrass across Western North America as a candidate, given its shallow roots, rapid drying, and contribution to wildfire intensity.
• •Single-letter DNA change prevents species extinction: Colossal's Australian partners identified that certain mammals eating toxic cane toads survive due to one amino acid change in a single gene. That same edit has been made in northern quoll cells in laboratory conditions. If validated in living animals, this single-letter genomic modification could prevent quoll extinction and enable the species to consume cane toads — an invasive amphibian currently lethal to all native Australian predators that encounter it.
• •Human embryo genetic selection is already commercial: Multiple companies now offer deep genomic sequencing of IVF embryos, providing prospective parents with ranked predictions for traits including height and cognitive scores, not just disease risk. Height heritability in Northern Europeans traces partly to Yamnaya steppe people who migrated into Europe approximately 4,700 years ago, demonstrating that population-level trait distributions already reflect historical selection events — making current commercial embryo selection a continuation of processes that have always shaped human genetics.