Length of Human DNA in bytes


Recommended Posts

It's a very valid question and I have a very good reason for asking it. I will post more about my reasoning in another tread later on.

I had a look on Yahoo Answers and I read this: http://answers.yahoo.com/question/index?qi...25135324AApHMlU

I wonder if the Human Genome Project have the full file for us to download? If they RAR it up, the file should compress nicely for about 50MB.

Why would you need 12 bytes per base pair? There can be only four types of pairing (AT, TA, GC, CG). So, one pair can be encoded in two bits. [say, 00=AT, 01=TA, 10=GC, 11=CG]. Since a male human has 3080 Million base pairs in his DNA, it translates to 6160000000 bits or a little more than 734 MB. And for the human female with 3022 base pairs in her DNA, its 6044000000 bits or a little more than 720 MB.

That's of course assuming writing it in binary. Writing it in ASCII will be 8 times larger, and using Unicode will be two or four times larger than the ASCII text file. I don't know how well it will get compressed.

Edit: Wikipedia gets a big larger file size.

Edit 2: While I was calculating, virtorio beat me with the WP link.

Also, does that sequence include the RNA sequence too?

Central dogma of genetics: DNA -> RNA -> Protein. The short story is that RNA is transcribed from DNA (the genome).

So what are you asking for exactly? The entire genome as in all DNA contained in a human cell's chromosomes? Total coding regions? Total protein-encoding genes? Total DNA and products such as mRNA, siRNA, miRNA, tRNA, rRNA?

Say, 00=AT, 01=TA, 10=GC, 11=CG

If you're willing to simplify and assume Watson-Crick base-pairing, then you may as well take only the total genome length (as opposed to total genome length x 2 to account for dsDNA), since A "always" pairs with T, and C "always" base pairs with G. If you want to get more complicated, you will have modified bases, ranging from methylation, acetylation, and a few other base analogues, as well as non-W/C pairing.

Also note the presence of mobile DNA elements such as (retro)transposons. They can move or replicate and insert themselves back into the genome. You also get all sorts of other crazy events like site-specific recombination that can result in duplication or excision of a stretch of DNA from a replicating chromosome, so really, the take-home message is that it is a freaking miracle that anything manages to live.

the take-home message is that it is a freaking miracle that anything manages to live.

lol, this^

I think there are quite large issues with the HGP (or anyone else) publishing somebody's genome, I doubt they'd allow any Joe Bloggs to download it.

edit: "All genome sequence generated by the Human Genome Project has been deposited into GenBank, a public database freely accessible by anyone with a connection to the Internet."

This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.