Cavalcade of Mammals

The first papers from the ENCODE Project: ENCyclopedia Of DNA Elements came out in Nature and Genome Research this week. The project’s goal is to understand non-coding DNA seqeunces by focusing in on 1% of the genome with multiple experimental approaches.

It’s not getting described well in the press. This news article is quite giddy: link

DNA study challenges basic ideas in genetics
Genome ‘junk’ appears essential

By Colin Nickerson, Globe Staff | June 14, 2007

A massive international study of the human genome has caused scientists to rethink some of the most basic concepts of cellular function. Genes, it turns out, may be relatively minor players in genetic processes that are far more subtle and complicated than previously imagined.

Among the critical findings: A huge amount of DNA long regarded as useless — and dismissively labeled “junk DNA” — now appears to be essential to the regulatory processes that control cells. Also, the regions of DNA lying between genes may be powerful triggers for diseases — and may hold the key for potential cures.

The details are interesting for biologists but not super surprising. Here’s an overview of non-coding DNA:

2% of the human genome encodes proteins. A difficult problem is to figure out what the the remainder, the non-coding DNA, does. About 1% is clearly involved in regulating protein expression, and a few additional percent have other known functions. An additional 2% are conserved non-coding sequences, DNA that has kept its sequence for tens of millions of years. The function of most of these conserved non-coding sequences is not known but their DNA has been retained because it has a function.

While it is hard to prove a negative, most of the remaining 90+% is likely non-functional (’junk DNA’) or very weakly functional. Some bits are likely spacers the sequence of which is unimportant. About 1/2 the human genome is transposons, bits of DNA that have copied themselves over and over (it is like the genomic inbox is full of spam).

The most interesting finding is that some non-coding non-conserved DNA is functional. Exactly how this works biologically is not clear–so very new, very cool. Ultimately, this adds another 1-2% to the functional non-coding class of DNA.

So it is interesting to hear that a new function for a bit of the non-coding DNA has been discovered–there will be a lot of cool new biology working out what bits of this do. Still, most of the genome is junk DNA.

The amount of junk DNA in a genome can vary a lot. There are some frogs and salamanders with genomes ten times larger than humans. These animals likely have about the same number of genes as humans. They are thought to just have a *lot* more junk DNA.

The recent development of the 1TB hard drive reminded me of my first thoughts about the end of the hard drive expansion. Back when the 1 GB drives came out I calculated my total lifetime media need. It came out under 10TB and half of that was video:

1TB 200,000 books at 5MB/book (every book I’ll read or know about)
1TB 200,000 songs, 5MB/song (every song I’ll ever hear)
3TB 3000 movies, 1GB/movie (every movie I’ll watch or desire to watch)
1TB 300 TV show seasons, 3GB each (ditto)

Personal records:
1TB 500,000 photos, 2MB/photo (every photo I’ll take or see)
1TB 2000 hours home video, 500MB/hr (fun to shoot, but who will watch it?)
0.001 TB (every word I’ll write, email, publish)
0.1TB 1GB/year (saved email and internet cruft, LOL cats, etc.)

It sums up to 8TB/life. And indexes, you definitely need indexes, so say 1TB for indexes.

This will expand as high res video/photos phase in but even that won’t push this out much beyond 100TB.

300 TB would allow you to record your entire life in video for 16hr/day for 100 years at 500MB/hr (one camera angle). No compression for time spent watching your self-video is assumed. :)

Saw this colored peg low res image and was reminded again I want to do something similar with black and white ping pong balls. I would work up a Lego bot to drop the balls in.

And then I had a cooler idea–I’ve seen a programmed water drop device used to make images. A related idea would be to drop water onto piano strings. The drops would strike the strings and make sound instead of images. The drops would fall a distance so you would see the music visually–see the notes coming before they hit.

Thinking more, there’s a good chance this wouldn’t work–the water drops wouldn’t have enough force. Worth a try though. :)

It was announced today that the full genomes of James D. Watson (and Craig Venter, though for much more than $1,000,000) have been sequenced. The NYT article had this bioethics blurb:

Dr. Watson and Dr. Venter are both taking a considerable personal risk in making their genomes publicly available. As is probably true for everyone, their genomes are likely to contain mutations that could lead to disease, revealing possibly unfavorable information about themselves and their relatives.

For Venter this is clearly untrue. He’s rich and can self insure with no problem. Likely Watson has enough dough this isn’t a risk either. For their poorer relatives, yes there is risk. I don’t think the writer of the NYT article, Nicholas Wade, gave this any thought–genetic knowledge insurance/employment risk is a standard story line, and the writer plugged it into this article.