Over the decades there have been many experiments involving random sequences, e.g. testing ATP-binding (e.g. Keefe & Szostak 2001), and evolving promoters (Yona et al. 2018), for the latter a good fraction evolved to match the wild type; I think this new study is the first to test the adaptive potential beyond promoters. I've also previously shared: Sahakyan et al. (2025): New study: In silico evolution of globular protein folds from random sequences : evolution. Chou et al. (2026): New study reconstructed the evolutionary history of a de novo gene that emerged in the common ancestor of simians : evolution Without further ado: Significance How new genes arise and gain function is a central question in biology. New genes can evolve from nongenic DNA, yet their adaptive potential remains unclear. Here, we use millions of (semi-)random sequences as experimental models of emerging genes and find that thousands confer phage resistance in Escherichia coli.…