2 ms·

I wonder if you could re-assemble the broken (due to decayed bonds) strands to the most likely sequence, given DNA sequence statistics similar to n-gram languag

• bognition 5y ago
  Yes you can, this is how most genomes are sequenced these days. You break the DNA up into a lot of smaller segments that are sequenced in parallel, you then combine them into the larger unified sequence. https://www.genome.gov/genetics-glossary/Shotgun-Sequencing https://www.genome.gov/genetics-glossary/Shotgun-Sequencing You cannot unfortunately do this for dinosaur DNA. The longest human chromosome has 250M basepairs. With a half life of ~500 years it only takes about 14K years for the chromosome to break down into individual nucleotides. 2^(14000/500) is roughly 170M
• tgb 5y ago
  Your body does this itself. You have two copies of every chromosome and they’re practical identical. A break in one will be repaired by using the other chromosome. Similarly, even a base change in one strand of one chromosome will be repaired based off the other strand (with a chance that the wrong strand is chosen tho). DNA repair mechanisms like these are exploited by CRISPR and other gene modification strategies: break the part you want to change and then trick the repair mechanism into making the modification you want by providing it with an appropriate template to work off. (Not all CRISPR based techniques use this.)
  • dekhn 5y ago
    https://en.wikipedia.org/wiki/Homologous_recombination https://en.wikipedia.org/wiki/Homologous_recombination