Cavalcade of Mammals

There are several dietary products that try to minimize the calories absorbed in the GI tract. One was a non-absorbable fat, Olestra. The fat just runs through the GI tract with a side effect of diarrhea and occasional worse effects. There are also drugs that keep fat from being absorbed: Xenical, a drug that inhibits intestinal lipase and slows the breakdown and absorbtion of fat. Alli is a low dose of the same drug available OTC.

There are also several types of fiber sold as fat-trappers. It’s not clear whether they work, but they also have the same side effects as fat blockers.

There are other ideas that have been tried. Stimulants like amphetamines work fairly well with some well known side effects. So far, drugs that mess with the regulation of appetite haven’t worked well–the regulation has too many redundant pathways.

Rather than blocking fat, how about using enzymes to breakdown either sugars or fats? The simplest approach would be to use enzymes, I’m sure suitable ones could be found, grown in E. coli, isolated and taken as gel caps. To keep the proteins from getting denatured by stomach acids, a coated or time release capsule could be used. A second step would be to engineer the enzymes to resist digestion–synonomous substitutions and so on. Another possibility would be to express the enzymes as secreted proteins in a gut bacterium–one of the ones that mainly lives in the jejunum. The bacteria could be ingested in pills the way probiotics are.

A more difficult implementation would be to non-protein enzymes to digest sugar or fat. This would be harder to engineer but likely more effective.

Which enzymes? Well, that would take some study. Likely a two or three would be needed to break down the metabolite and then waste any ATP formed.

I’ve put up a web site for exploring L-system images, L-system Iterator.

The snowflake shape is only one example of the pictures that can be drawn this way.

L-systems are simple iterated drawing rules. Simple rules for turning and drawing put together in this way make quite interesting and complicated patterns. The ones shown above are well known. From the the left, the Koch snowflake, the Sierpinski triangle, a kolam-like image, and a plant-like image. On the second row, the Heighway dragon, the Hilbert curve, and another plant.

The Heighway dragon has many interesting properties–for example, it can be tiled over the plane.

Some iterated objects are fractals–the Koch snowflake, Sierpinski triangle, and Hilbert curve are famous simple fractals.

L-systems can be quite complicated. The systems modeled on my web site use a single rotation angle and only one line width. More complicated models can make surprisingly realistic plants. Prusinkiewicz and Lindenmayer (the L in L-system) have developed detailed plant models.

The web site is based on the Perl code I wrote for my Biomorph evolution/selection web site. The images are generated using Postscript to draw the L-system, and then the ImageMagick convert program to change it to a PNG image. Images are given a file name that describes the L-system, effectively caching the image. The Prototype Javascript library is used to assist in making the popup boxes.

Each L-system variant has two changes from the current L-system. Some logic is used to keep the L-system in the same family–if there’s no Y equation, one isn’t added. Existing equations are grown or shrunk but not dropped. These images can take much longer to generate than the Biomorph images, so a number of limits are placed to keep the L-system from getting too complicated or taking too much CPU time.

The hardest part of the Postscript was getting the images scaled and centered appropriately. The images can extend in any direction and some are large, others small. The centering code generates the image twice, once to find out its dimensions and then a second time scaled and centered. Here’s the code to record the dimensions of each part of the image. It gets called before each stroke operation.

/max_path {
gsave initmatrix pathbbox grestore



ur_y false eq { /ur_y exch store } { dup ur_y gt { /ur_y exch store } { pop } ifelse } ifelse
ur_x false eq { /ur_x exch store } { dup ur_x gt { /ur_x exch store } { pop } ifelse } ifelse
ll_y false eq { /ll_y exch store } { dup ll_y lt { /ll_y exch store } { pop } ifelse } ifelse
ll_x false eq { /ll_x exch store } { dup ll_x lt { /ll_x exch store } { pop } ifelse } ifelse
} def

The ‘initmatrix’ command is required to reset things because of all the rotation operations.

The code for the site is linked on the L-system iterator home page.

Update: Added color variation as an option. And a reverse direction primitive.

Also, the code now runs under mod_perl.

Note for mod_perl users–mod_perl 2.0 has no way of handling alarms. select() doesn’t work either as a way of timing out pipes. The only usable method is prepending commands with ‘ulimit -t secs’ and letting the shell limit the system process.

To make the split color B&W images I used these ImageMagick commands:
convert -size 150x150 tile:color.png tile:bw.png ../temp/mask.png -composite split.png
using a half black, half white split image as the mask.

Then added the split line using:
convert -size 150x150 -fill white -stroke black -draw "line 0,0 150,150" split.png split_line.png

Here’s an idea for a game. It would be like the kid’s game Memory, where cards are turned over and matches are taken off the playing area. Instead of using identical pictures, images of different animals or plants would be used. Any pair could be matched by a player. The play would be quite similar to standard Memory.

The idea would be to match organisms by evolutionary similarity. So scoring would give maximum points for animals of the same species, next most for same genus, fewer points for animals in the same order, and no points for creatures in different phyla. The easiest implementation would be as a computer game with the computer dealing with scoring. Alternatively cards could be made the lineage described on the back. Each classification category could be displayed a different color or with a different symbol and the first/highest point matching lineage symbol give the points for the match.

This would make the play interesting as any pair could be matched but the player would have to decide if a pair was good enough to pick up or to wait for a better and higher scoring pair next turn.

The design aspect of picking a card set could make an easy set or a hard set, and two aspect of the choice would affect this. First, if animals fall into close pair groups that are distantly related (two parrots, two foxes) then the set would be easier. Having graduated and overlapping groups of cards make the set harder (dog, fox, skunk, weasel, otter, raccoon). Also, how much the player knows of these animals and their relationships can make a card set easier or harder. Some groups are obvious–birds, whales, bats–while other animals are either not as well known (i.e., coatis) or don’t have an obvious lineage (i.e., wolverine). And all these examples are mammals. Invertebrates would make a ridiculously hard game! So sets for kids could be easy and moderate to hard sets can be created.

Here are three game sets:
Mammals, butterflies, and marine invertebrates.

I have thought for years that it should be possible to regrow teeth. Teeth should be one of the easiest body parts to regrow. It seemed likely that the tooth bud, once formed, would receive signals from its local environment and grow into the correct type of tooth and emerge into place. That’s what happens during normal adult tooth development. So generating a tooth bud looks to be the key step. And indeed, in the past few years there have been reports of progress from research in this area. See this news article and this paper from the Yelick lab in São Paulo, Brazil.

But much easier than growing teeth should be killing tooth buds. Specifically, if the buds of wisdom teeth were killed then the painful, expensive surgery to remove wisdom teeth could be avoided.

Tooth buds form during fetal development. Wikipedia has a detailed overview. Wisdom teeth don’t begin to calcify until a person is 7 to 10 years of age. It should be fairly easy to kill the tooth bud at early stages. An injection into the tooth bud of a localized cytotoxin, either a general one or perhaps one specific to dividing cells would kill the stem cells that form of the core of the tooth bud. A toxin dose that kills cells within a 1-2 mm radius of the injection site should be effective. The gums will heal up and then the tooth bud will be gone. The dentist should be able to pick the injection location based on the expected eruption site. Inspection of x-rays may help pinpoint the bud location. A jig could be used to precisely position the needle tip.

Googling briefly I don’t see any other mention of this idea. It would be easy to test experimentally in animals if one can be found with late enough tooth development.

What would be required to set up an inexpensive system for hobbyists to experiment with biology? Consider PCR for example. PCR requires heat stable polymerase, primers, nucleotides, buffer.

The DNA polymerase is easy to purify from E. coli carrying the plasmid. Grow bacteria containing plasmid expressing Taq DNA polymerase, boil, spin down denatured proteins, and you are left with the DNA polymerase.

Primers can be bought inexpensively–$0.35 per base, a pair of 18-mers cost less than the shipping. Though they are inexpensive only if one set gets used repeatedly.

The cost of buffer (NaCl, MgCl₂, Tris) is negligible.

Nucleotides are expensive up front, $150 for a set of dNTPs (dATP, dCTP, dGTP, dTTP), but this works out to about $0.06 per 50 ul PCR reaction.

Can nucleotides be prepared by a hobbyist? Nucleotides are easy to obtain-DNA is a major constituent of cells and is easy to purify. DNA + DNAase = dAMP, dCMP, dGMP, and dTMP. How can the trinucleotides be regenerated?

One route is to do it enzymatically using
polyphosphate:AMP phosphotransferase (PPT) and adenylate kinase (AdK) with polyphosphate (polyP) as the energy source (Resnick and Zehnder, 2000). It is not clear how the trinucleotide product would be separated and purified. Presumably different enzyme pairs could be used to regenerate the other dNTPs from the monophosphates.

These other enzymes could be cloned in E. coli expression vectors and purified either by tagging them with His₆ and using a Ni or Co resin. Or by cloning heat-stable isoforms from one of the extremophiles and using a one-step boiling purification like that used for Taq polymerase.

Update: Bochkov et al., 2006 describe a method of preparing dNTPs from digested DNA. DNA is digested with DNAase and Nuclease S₁. DNAase chews DNA into show oligonucleotides and the nuclease breaks them down to single dNMPs.

Then a crude extract of E. coli is prepared that contains the kinases to convert dNMPs to dNTPs along with the acetokinase. The kinases use ATP. ATP must be regenerated, and this is done using acetokinase with acetyl phosphate ($30/g) as an energy source. Combined dNMPs were converted to dNTPs with at least 86% regeneration and separated from reactants by chromatography on a Dowex 1×2 anion exchanger. The conversion was followed by thin-layer chromatography.

For PCR it may be possible to use a crude purification of nucleotides, but purification protocols would need to be developed and tested.

Phil Agre ran the Red Rock Eater mailing list for many years, it was active during the 90’s. I think I started reading it in the late 90’s. The emails ran down to a trickle in 2002, in part I think due to Phil getting a permanent position at UCLA. The mailing list archive indicates it ran up to Jan 2005.

Phil Agre’s mailing list in essence was one of the first blogs. The content was mainly links and commentary by Phil with occasional longer essays. He was one of the best people thinking about what the internet could be used for and how it was changing the world.

After 2002, the Red Rock Eater list went into abeyance then seemed to have stopped for good. Phil Agre seems to have dropped off the net. In this comment thread a UCLA student says he was ill. I fear it is very serious to keep him off line so long.

His essays, “Advice for undergraduates considering graduate school” and “How to be a leader in your field” are internet classics.

Here is an essay I found interesting titled What Is Conservatism and What Is Wrong with It?

Update: He’s literally missing. This site is run by friends looking for him.
Update 1/31/09: UCLA police talked with him. He’s alive, though not well.

It would be useful to attach a static electricity generator on the end of a long thin rod to pick up small non-metallic things dropped behind bookcases or otherwise out of reach. Could run a band through the center of a thin rod, or easier yet just package a thin PVC rod with a piece of rabbit fur or nylon.

Cells become cancerous through a multi-step process. The cells pick up several mutations, each clearing a natural limit on cell division and usually increasing the rate at which the cells divide. By the time a person gets old their body has many pre-cancerous clumps of cells, and cancer occurs when one of the cells in one of the clumps picks up a final mutation and becomes fully cancerous.

Cancer has proven very difficult to treat, but perhaps it is easier to treat at the pre-cancerous stage. The idea would be to treat healthy people at middle age or later and kill most of their pre-cancerous cells. This would make the pool of cells that can develop into cancer much smaller and reduce the incidence of cancer.

Chemical chemotherapy drugs would be a poor choice for this–I expect they are not effective on slowly dividing pre-cancerous cells and these drugs are also damaging.

Instead, it may be possible to trigger apoptosis (cell suicide) in pre-cancerous cells. These cells are losing their differentiation and activating abnormal signaling pathways. They are likely stressed and may already be primed to undergo apoptosis. One of the organism’s anti-cancer mechanisms is to trigger apoptosis in pre-cancerous cells. The idea here is to supercharge this mechanism.

So the idea would be to treat the person with a cocktail of drugs that induces apoptosis by activating the apoptotic signaling pathways. The treatment should be strong enough to trigger a wave of apoptosis in the most susceptible cells clearing most pre-cancerous cells from the body. There would be some normal cells killed as well but they will be replaced by normal tissue processes.

The fracture of the US financial system gives me an idea for a new business model.

1) Raise some capital.

2) Use a tiny fraction to hire a few business/finance reporters, and have them find a company riddled with fraudulent business practices. How hard can it be–this stuff is common and not very well hidden. Look at Enron, Halliburton, the Madoff Ponzi scheme.

3) Then short the company’s stock, send the evidence to the appropriate regulatory and oversight agencies, feed the info to the press, and profit!

One of the coolest technologies around are the rapid prototyping technologies: CNCs, Laser scanning, and Stereolithography. I first ran into stereolithography at a UK campus fair soon after I arrived, I forget which one.

In stereolithography a 3D computer model gets formed into a actually plastic part using a process in which heat from a laser polymerizes a liquid. Each pass forms a layer, and layers of under 0.1 mm are added to build the complete part. This technology was developed in the late 1980’s and is now fairly common though very expensive. The machines are $50-150k, and shops will turn out a part for a few hundred dollars. Incredible stuff.

The University of Kentucky has a Rapid Prototyping shop with a Stereolithography machine on campus.

Two things make this tech expensive, the monomer plastic ($200+ per liter) and the machines. The monomer is magic stuff, 21st century chemistry, and can’t substituted.

Can a machine be made more cheaply? Perhaps. Instead of the UV laser and X/Y scanning optics, how about using a DLP and a UV bulb? The DLP (an array of computer controlled mirrors, the display tech for bright computer projectors) directs the UV light to the correct part of the work area, the monomer polymerizes, the part is lowered, and the cycle repeats.

Could this work? It seems plausible. Best as I can tell, the monomer needs to be heated 20-30°C to polymerize. This is done in ~100μs by the laser, so a UV bulb shining for longer might be able to polymerize the monomer. The UV laser (100mW typically) has an intensity equivalent to 10,000W shining on a small 2cm x 2cm work area, so a 200W or 400W UV bulb (commercially available) shining for 1/10 of a second should in theory work. Losses due to heat dissipation may make a longer light pulse necessary, but it would likely still be under a second if this will work at all. The monomer could be preheated to the highest temperature where polymerization doesn’t occur to help polymerization along.

The result would be a simple and cheap stereolithography device with a small working area (2cmx2cmx15cm)! It could be built for very little ($200 UV bulb, $100 old DLP projector, $200 lenses and misc parts), with the high cost of the plastic monomer the biggest limitation. A DLP chip with 1000×1000 mirrors would have a 0.02 mm resolution. If UV intensity required for polymerization isn’t limiting, at 0.075 mm resolution a standard DLP would give a 7.5cm x 7.5cm working area and allow larger parts to be built.

What is the cost in plastic monomer for the parts? About 15c to 50c for a small Lego size piece, compared to the 5c a block cost for Legos.

Update: I had a look around for cheaper & more accessible plastic monomer. Two turned up–rubber stamp photopolymer resin and screen printing photopolymer emulsion.

The rubber stamp photopolymer is inexpensive and would likely UV cure in under a minute (1s or 60s is hard to gauge). Unfortunately, the resin is quite viscous, 60,000 cps, quite thicker than honey. So likely too thick for adding a thin layer at a time.

Screen printing photopolymer emulsion appears more promising. It polyvinyl alcohol or polyvinyl acetate with diazo or SBQ crosslinking initiators. It forms a plastic that is fairly weak and not water resistant but which can be hardened with secondary UV curing and hardening agents. UV polymerization typically takes a few minutes (with a source of 1-2 W/cm²); a strong UV source should reduce the time to under a minute. These photopolymer emulsions typically have viscosities of 3,500 to 12,000 cps, much more suitable!