ApoB and LDL

Speaking of apolipoproteins (see below), a group from McGill University Health Center has found that screening for Apolipoprotein B (apoB)is a better measure for heart disease than simply cholesterol levels. They found that even in patients with low cholesterol, the apoB levels could still be high. One apoB molecule stabilizes a low-density lipoprotein particle (LDL). The more LDL you have, the worse the prognosis. So, apoB levels are a good way of seeing how much LDL you have.

Using Similarities to Find Differences

Thursdays and Fridays are big days for science news, since Nature comes out on Thursday and Science on Friday. PNAS seems to have new articles every day. News that is embargoed can now be published. So I scan the Table of Contents for interesting articles.

This one caught me eye: Phylogenetic Shadowing of Primate Sequences to Find Functional Regions of the Human Genome. Now I have no idea what the heck ‘phylogenetic shadowing’ is. A Google search finds 4 articles, all by the current authors, so this must be a neologism and will require reading of the article to determine what it means.

Help is on the way

Luckily, there is something from Eurekalert that gives a lot more detail of the process. With the title Scientists Find That Apes and Monkeys Provide Needed Help in Understanding the Human Genome it provides a nice background. Normally, we compare genomes of animals that had a last common ancestor maybe 100 million years ago. This is because it is much more difficult to find important genes in animals that are closely related. There are not enough differences to provide us any clues as to where important genes are.

A metaphor one of the authors uses is that comparing mice and humans is like looking at a go-cart and a car. Easy to detail the differences. But comparing baboons to humans, whose genomes are 95% similar, is like looking at a sedan and a station wagon. The differences are harder to detail, they are more subtle.

Comparing multiple genomes

If you look at humans and chimps, almost everything is the same, non-essential DNA from introns and the coding sequences from exons. But these researchers found a way. They simultaneously looked at the genomes of up to 15 non-human primates. The increase in information, coupled with filtering tools, helped tremendously. Now minor differences between each of them could be used to cumulatively identify the regions that coded for protein and those that did not.

They demonstrated the usefulness of this approach by examining a gene only found in primates, apolipoprotein A. This is a gene that may have some very important ramifications in heart disease. Using any other mammal than primates would have been useless. Using phylogenetic shadowing they were able to identify the regulatory regions that control expression of this gene. So, not only did they develop a novel approach for filtering the huge amount of information found today, they used it to examine a clinically relevant protein. Nice to see the combination of novelty and relevance.

A Rose by Any Other Name Would Smell as Sweet

Speaking of smells (see below), another paper from PNAS has some interesting insights into our sense of smell. Is the olfactory receptor a metalloprotein? is its title. I like titles as questions rather than declarative ones (i.e. The olfactory receptor is a metalloprotein). A smattering of Latin or a dictionary will tell you what olfactory means and Google will lead you to metalloproteins (proteins containing metals at their active sites)

Now the paper is kind of tough but, being the Information Age, there is a wonderful press release for it. One of the really unsung innovations today is the sometimes excellent scientific communication personnel at universities and corporations. This one provides some very useful facts. As does the main author’s website.

One is that there are 1000 olfactory genes. This represents 3% of the total number of genes we possess. For an animal whose sense of smell seems secondary, that is a lot of genes. And it appears that the presence of particular metal ions, especially zinc and copper, in some of these olfactory proteins is vital for their activity.

It appears like these proteins may have different activities depending on what metal is bound to them. This may also explain why a zinc deficiency can result in a loss of the sense of smell. All from one little paper. Sweet!

An Antidote to the Dew of Death

One of the great things about the free databases that are being created for scientific literature (such as PubMed) is that older papers will be as easy to find as recent ones. When I first started writing papers, it was very hard to really find and read papers that were more than 10 years old. The indexing services were just too crude and the field was moving too fast. So I really enjoyed it when I could include a paper from the 60s as a reference.

That is why I love finding old technology that can still be useful today. Researchers at the Indiana School of Medicine has revisited WWII era work that may have some real benefits fighting the tools of terrorism. Fear that the Germans would use a particularly nasty gas (Lewisite) against them, British biochemists developed some very potent countermeasures. The chemical, British Anti-Lewisite (BAL), that they came up with can work very well today against gases that Iraq does have the technology to produce.

As always, I search Google. Looking for anything at the author’s websites did not find anything. But, searching for BAL produced a wonderful article published by Notre Dame Magazine that was written by one of the authors of the current report, Joel Vilensky. Read the article. It is a nice history of one of the worst scientific ‘discoveries’ from WWI, the development of the antidote, and then its use to save children from lead poisoning or patients suffering Wilson’s Disease. The gas smells like geraniums. The antidote smells like rotten eggs. I really enjoy finding information like this on the Internet.

Wed, 26 Feb 2003 16:49:56 GMT

Fifty years ago Watson and Crick published a paper entitled
A Structure for Deoxyribose Nucleic Acid. Many publications are commemorating this, including Nature and the New York Times. It was one of the seminal papers of the 20th Century. But we only know that now, looking back. At the time it was only one of three short papers in an issue of Nature. The other two, one by Wilkins, Stokes and Wilson, and the other by Franklin and Gosling, actually provide some real data to discuss.

Watson has said that they were able to work in relative obscurity for several years before others really started to follow this model. The popular press did not report anything for almost a month.

Reread the paper with an open mind and you may be unimpressed. They propose a model with little documented proof and with many obvious problems. In fact, the model that this paper discussed did not immediately revolutionize the world . There were lots of good reasons for the slow acceptance, but, as with any good hypothesis, it presented a framework for determining its value or not.

As often happens in science, the really great papers, the ones that lead to simplifying descriptions of the living world, slowly reveal their importance. Models must be substantiated with scientific proof. Today, with the Internet, it is too easy for the opposite to occur. Everything gets hyped and there is a press release for every little paper written. Scientists, as well as everyone else, will have to gain better filtering mechanisms to deal with this. The Faculty of 1000 from BioMedCentral is one such attempt. I wonder how Watson and Crick’s original paper would have fared?

Sat, 22 Feb 2003 22:03:11 GMT

As we gain more knowledge of the different genomes, we may be gaining some understanding of what early life was like. In particular, it appears that horizontal gene transfer may have been a very important process for all life.

In this case, two of the most important cellular pathways, photosynthesis and nitrogen fixation, may have come about from the mixing of several different genes from different organisms. This group as ASU also found a group of similar genes that code for proteins that do neither photosynthesis nor fix nitrogen. These genes appear to be relics from a very ancient time and may be the ancestral genes for both photosynthesis and nitrogen fixation

It is still a mystery just what these proteins do. But, they found a bacteria containing these ‘old’ genes. Methanococcus jannaschii. It lives in concentrations of hydrogen cyanide that would kill other organisms. But cyanide would make a nice source of nitrogen and it would be plentiful in the early atmosphere.

The first college professor I ever got to work with, Irwin Spear, told the story of finding some bacterial growth in a concentration of cyanide. Being young, he tossed it down the drain before wondering just what sort of bacteria could grow in a solution that would kill everything else. Perhaps it was this bacterium or one similar,, trying to find a little bit of the primordial world in a small bottle in a laboratory refrigerator.

The ability for any researcher to access important work at any time is one of the fundamental changes that this age is providing. Now, this access may well become free. BioMedCentral has a very strong mission to provide just this sort of journal. A recent article in one of their peer-reviewed, open access journals serves as a useful example.

Entitled Severe Anaphylactic Reactions to Glutamic Acid Decarboxylase (GAD) Self Peptides in NOD Mice that Spontaneously Develop Autoimmune Type 1 Diabetes Mellitus. (Don’t you just love titles from research papers). What it discusses is the use of small peptides to help ameliorate the development of diabetes in a special strain of mice. This is a very hopeful therapy that is being used in some recent clinical trials in humans.

This paper indicates that using peptides can sometimes cause more harm than good. In this case, all the mice that got the therapy died. Pretty important stuff and something that anyone can read without having to pay a tribute to a commercial publisher. Of course, it is 16 pages of pretty dense material but that fact that you COULD read it is what matters.

Fri, 21 Feb 2003 00:54:16 GMT

There has been much written about the new age we are entering, the new paradigms we are encountering. The ability to rapidly form adaptive networks permits us to find novel ways to approach and to solve a wide variety of difficult problems. Information wants to be free. But are these more that New Age parables and sloganeering?

Not in my opinion because I have lived them. I have seen the changes that new technologies have made on the study of life. They permit us to answer questions that were impossible just a few years ago. And, as science always does, each answer reveals ten more questions, requiring more investigations.

When I started as a graduate student, it took about 2 days to sequence a DNA fragment 10 bases long. Now we are approaching a time when one machine can sequence one million bases in a single day! Data from genomics and DNA sequencing is doubling every 6 months. Other technologies may generate 100 times more data and require 1000 times the computing power. While allowing us to do so much more today, these same technologies are creating some real difficulties.

Too much noise, not enough signal. We need new ways to attack this problem. The old ways are no longer sufficient. The groups that discover and foster novel approaches to these problems will be much more successful than those that use last century’s methods.

Some of these approaches are simply extensions of the way science has been increasingly done: transparency of information, openness in its dispersal, collaboration, peer review, reputation. The tools of the Information Age simply reduce the friction of these approaches, making it easier to generate data – producing a mountain of information but also providing us with tools that can extract knowledge from this mountain.

So, what will Living Code be about? Well, it will have a definite point of view. Science is so often portrayed as a set of dull principles, yet many of the scientists I know are anything but dull. They are excited, curious and open to novelty. In short, creative human beings. I want to make sure that those aspects come through.

Collaboration and openness create knowledge from information. I will look at ways that these principles help us gain a better understanding of the natural world. I will examine the knowledge that is being generated, and some of the questions that are raised. I will add my commentary and perspective, in a style that will, I hope, be thought-provoking.

Biotechnology – the problems it faces and the solutions it will create – serves as a surrogate for many other arenas that are rapidly changing. Several of the other weblogs at Corante address these areas. Understanding how biotechnology solves its difficulties, how it succeeds in dealing with the torrent of information, may help us understand how to solve problems in these other areas. I hope so. It is why I am writing.

A Protein Plot

The Ramachandran plot was developed over 30 years ago to help describe how 2-dimensional structural elements (alpha helix, beta sheet) of a protein could be defined. Along with more recent developments, it forms a backbone for many of our attempts to determine protein structure. Now a similar approach has been published by Sung-Hou Kim and his colleagues that characterizes how the 3-dimensional elements of proteins may be categorized.

This work is based on the databases of protein structure that have been generated over the last 50 years. Looking at a subset of 500 protein structures, they examined the differences between each of them. This required a huge number of calculations and was very computer intensive. But when they were done, they had been able to calculate some things that we had only suspected before.

As you move out along any of the axes, the proteins get larger and more complex. There are 4 main types of proteins. The data seem to indicate that 3 of the 4 have a common origin (perhaps the primordial protein) and that the fourth is a much more recent development. And, the graph that was generated numerically matches quite well with a database of structures that has been hand-generated.

So, in the future, proteins can be placed into the proper group on this plot and we will know quite a bit about it. I would be really interested to know which proteins are close to the origin of the plot. (a Kim plot?).

Thu, 20 Feb 2003 02:12:15 GMT

There is another Ebola outbreak in Africa and new avian influenza cases in Hong Kong. Ebola is incredibly efficient at converting human beings into virus particles. The genome of Ebola has been sequenced. It is an extremely simple virus with 7 genes that code for 8 proteins. Several of these attack the walls of blood vessels, generating the major destruction this virus wrecks. A mystery since its first description is Where does Ebola hide between epidemics?. It seems to kill almost every primate it infects. We found out about the latest epidemic first by reports of deaths in mountain gorillas. All of our tools have not been successful at solving this problem, but they are making it more likely that people will survive the disease.

The new outbreak of avian influenza is also particularly noteworthy. In 1997, such an outbreak resulted in very high mortalities in chickens and a pretty high (30%) mortality in humans. Luckily, the outbreak was contained. A real difficulty is that our methods for developing vaccines are very old. Flu vaccine is still generated by using chicken eggs. If the virus kills chickens, we have no way to make a vaccine. Luckily, newer technologies are being investigated. They also allow us to more rapidly determine which vaccine needs to be used. Constant vigilance is needed in our battle against infectious diseases but new technologies, along with human creativity should keep us ahead of the game.