Slashdot Mirror

"If any (non-trivial) treatment consisting of specifically folding proteins is found, then there will be exactly one way to produce said drug : genetic manipulation. Only a genetically modified cell will be able to produce those custom proteins."

This is not necessarily the case. We can predict with reasonable accuracy (about 80%) what sort of secondary structure (alpha helix, beta sheet, or coil, basically) a protein will have based solely on the protein sequence. For example, if you were to put the protein sequence lkgtlgqdvidirtlgskgvftfdpgftst, into Jpred you get this prediction:
LKGTLGQDV IDIRT LGSKGV FTF DPGFTST
------------------EEEEE-------------EEE
Where E is the code for beta sheet, and the dash is coil. Jpred is smart enough to also search through the Protein Data Bank and see if that sequence belongs to a protein that has already had it's structure solved, which is the case here. I told Jpred to go on with the prediction anyway, and by comparison to the known protein structure iIt doesn't do horribly bad, correctly predicting that most of this sequence is coil, getting close with the first bit of sheet secondary structure, and misinterpreting the second sheet bit when it should be another type of structure (a turn--several different ones are defined but make up only small amounts of structure, being largely transitions between the other secondary structure types). However as the example sort of hints at, protein structure can be squishy. If your protein is an enzyme, then it must bind substrate, catalyze a reaction, and release product. There will be some structural change that accompanies this, meaning that small molecules can have an impact on protein structure. Sometimes but not often this can involve radical change. I knew a guy a few years ago who designed a protein, such that he could control the secondary structure of a short stretch by the addition of a small molecule. If present, part A was helix, part B was coil. If absent, part A was coil, part B was helix. I'm running out of time, but there are also examples in nature of a protein in the course of it's biological function that has parts swap between sheet and helix structure. Lastly, we come to diseases caused by misfolding proteins. The mutations are often small ones, that disrupt proper folding of structural proteins. It may be possible to treat disorders at least to some extent by introducing a small molecule that will favor the properly folded state as opposed to the disease-causing improperly folded protein state (I'm thinking of Lou Gehrig's disease aka amyotrophic lateral sclerosis, ALS). Such a small molecule probably can't do anything about existing misfolded protein stuck in a plaque in the cell, but the small molecule might interact with protein as it is being made or before it joins the plaque and prevent an increase in the plaque size. It is at least conceivable that a conventional drug treatment could affect this part of ALS and halt or slow the progression of the disease.

It's a student (undergraduate I presume) project, so I expect the funding was zero.

This page indicates that she is a postgrad.

Apart from the inappropriate use of the 'PC' term, the article doesn't actually state what type of architecture the processor is.

This far more interesting paper on it, states that it has a MIPS processor.

There is no 'hardcore' surveillance going on. These are weather and climatology satellites. They're using AVHRR and MODIS data to model when conditions are ripe for insect outbreaks that can carry malaria. The best resolution in the study appears to be 250 meters, so it isn't capable of picking out individual houses, let alone spying on people.

It's sent to the DoD because they have a centralized database that combines these 14 satellites, coregisters them with each other, and turns them into one product. The DoD is making scientific research much easier, here, by letting researchers access this combined database. That's why it's newsworthy.

Some of my colleagues (e.g., Vittorio Cristini) have been modeling the potential benefits of nanoparticle drug delivery for a couple of years now. As has been known for some time (e.g., see papers from R.K. Jain), the blood vessels that grow to supply tumors with nutrients (the tumor-induced neo-vasculature) are different than regular, non-pathological vessels. They tend to be more tortuous and leaky, with larger holes than regular vessels.

This is where the nanoparticles come in: one can design nanoparticles that encapsulate cancer drugs in particles that are too large to exit normal blood vessels but can pass through the leakier, tumor-induced blood vessels. This naturally targets cancerous tissues.

However, there are other issues to consider. Due to the high pressure inside tumors (due to the rapid proliferation of cells within a confined area, among other factors), along with the leaky vessels, blood flow can be very poor inside a tumor, and so while the drug may be targeted toward and delivered to the tumor, it may not actually penetrate very far into the tumor. Some great work has been done by Steven McDougall, Sandy Anderson, and Mark Chaplain in this area. In particular, look at their DATIA (dynamic adaptive tumour-induced angiogenesis) papers.

One way around this (suggested by R.K. Jain and Vittorio Cristini, among others) is to use targeted anti-angiogenic therapy to prune out the worse blood vessels and improve flow within the tumors, thereby also improving drug delivery and penetration.

Lastly, on the therapeutic aspect of blocking up tumor blood vessels with the nanoparticles, the work we've done (see this paper, which will appear in the Journal of Theoretical Biology soon), indiscriminately cutting off the nutrient supply to a tumor can increase tumor invasiveness by increasing morphological (shape) instability. (See some of the animations here.) So ironically, while more tumor cells may be killed, those that remain may spread farther and initiate new tumors. Given that hypoxic tumor cells are more likely to be resilient to further treatment (e.g., hypoxic breast cancer cells), this is a problem worth keeping in mind when planning anti-angiogenic therapy.

If you're interested in these topics, please do check out the paper above. (You can also download it at my website without any special memberships.) Even if you don't like it, we have a lot of references you may find handy. -- Paul

Some of my colleagues (e.g., Vittorio Cristini) have been modeling the potential benefits of nanoparticle drug delivery for a couple of years now. As has been known for some time (e.g., see papers from R.K. Jain), the blood vessels that grow to supply tumors with nutrients (the tumor-induced neo-vasculature) are different than regular, non-pathological vessels. They tend to be more tortuous and leaky, with larger holes than regular vessels.

This is where the nanoparticles come in: one can design nanoparticles that encapsulate cancer drugs in particles that are too large to exit normal blood vessels but can pass through the leakier, tumor-induced blood vessels. This naturally targets cancerous tissues.

However, there are other issues to consider. Due to the high pressure inside tumors (due to the rapid proliferation of cells within a confined area, among other factors), along with the leaky vessels, blood flow can be very poor inside a tumor, and so while the drug may be targeted toward and delivered to the tumor, it may not actually penetrate very far into the tumor. Some great work has been done by Steven McDougall, Sandy Anderson, and Mark Chaplain in this area. In particular, look at their DATIA (dynamic adaptive tumour-induced angiogenesis) papers.

One way around this (suggested by R.K. Jain and Vittorio Cristini, among others) is to use targeted anti-angiogenic therapy to prune out the worse blood vessels and improve flow within the tumors, thereby also improving drug delivery and penetration.

Lastly, on the therapeutic aspect of blocking up tumor blood vessels with the nanoparticles, the work we've done (see this paper, which will appear in the Journal of Theoretical Biology soon), indiscriminately cutting off the nutrient supply to a tumor can increase tumor invasiveness by increasing morphological (shape) instability. (See some of the animations here.) So ironically, while more tumor cells may be killed, those that remain may spread farther and initiate new tumors. Given that hypoxic tumor cells are more likely to be resilient to further treatment (e.g., hypoxic breast cancer cells), this is a problem worth keeping in mind when planning anti-angiogenic therapy.

If you're interested in these topics, please do check out the paper above. (You can also download it at my website without any special memberships.) Even if you don't like it, we have a lot of references you may find handy. -- Paul

So, while the mathematical model of growth might represent some predictive value, it certainly will not effectively model new developments, such as the above, when they are found.

There's still plenty of value to be found in higher-scale models. (e.g., how the tumor as a whole interacts with the microenvironment, how proliferation-induced pressure turns off the vasculature and prevents drug delivery, how oxygen and glucose delivery throughout the tumor and the microenvironment affects the patterning of hypoxic and necrotic cells, which, in turn, affects angiogenesis and matrix degradation) Cancer is a multiscale problem, with interaction between all the scales. Focusing on one scale alone (molecular or tissue-scale) likely will not solve the entire problem.

In fact, developing a good tissue-scale model is a natural step toward creating a multiscale model, where molecular- and cell-scale dynamics affect the growth parameters that govern tissue-scale behavior. (Similarly to how the behavior of individual molecules leads to things that can be averaged at larger scales, like heat, viscosity, etc.) First, you fix the parameters and neglect the small-scale dynamics to figure out the large-scale behavior. Then, you model the small-scale dynamics and learn how to couple them to the previously-fixed parameters.

On a tangent, I also worked with "Sandy" Anderson in a different U.S.-Scottish collaboration. He's a really great guy, as well as Mark Chaplain and Steven McDougall. And Sandy is a pretty incredible cook! :-) -- Paul

So, while the mathematical model of growth might represent some predictive value, it certainly will not effectively model new developments, such as the above, when they are found.

There's still plenty of value to be found in higher-scale models. (e.g., how the tumor as a whole interacts with the microenvironment, how proliferation-induced pressure turns off the vasculature and prevents drug delivery, how oxygen and glucose delivery throughout the tumor and the microenvironment affects the patterning of hypoxic and necrotic cells, which, in turn, affects angiogenesis and matrix degradation) Cancer is a multiscale problem, with interaction between all the scales. Focusing on one scale alone (molecular or tissue-scale) likely will not solve the entire problem.

In fact, developing a good tissue-scale model is a natural step toward creating a multiscale model, where molecular- and cell-scale dynamics affect the growth parameters that govern tissue-scale behavior. (Similarly to how the behavior of individual molecules leads to things that can be averaged at larger scales, like heat, viscosity, etc.) First, you fix the parameters and neglect the small-scale dynamics to figure out the large-scale behavior. Then, you model the small-scale dynamics and learn how to couple them to the previously-fixed parameters.

On a tangent, I also worked with "Sandy" Anderson in a different U.S.-Scottish collaboration. He's a really great guy, as well as Mark Chaplain and Steven McDougall. And Sandy is a pretty incredible cook! :-) -- Paul

this is nothing new: it started before the WWI and now there are dozens of companies, universities or hobbyist doing it. It is called: "content analysis", "data mining", "discourse analysis" etc. There is a legend that sais that British intelligence managed to predict quite acurately airstrikes on England based on content analysis of Goebels' radio speeches. Take a look at this links if you are interested. Bibliography of Content Analysis Listings from Communication Abstracts, 1990-1997 Content Analysis Resources web site Text Analysis Info Page - all on text analysis and related topics The discourse analysis page of AI Topics Centre d'analyse des politiques publiques (CAPP) Département de science politique, Université Laval The Center for Social Research Methods: not necesarily content analysis, but it's good to take a look at Research Methods Knowledge Base The Annenberg School for Communication Web Concordances at the English Department of the University of Dundee Companion Website for the book Word Frequencies in Written and Spoken English: based on the British National Corpus Journal: Language Awareness; has some free issues/articles. The General Inquirer Home Page Journal of Second Language Writing Writing Guides: Conducting Content Analysis at Colorado State University; with a nice adnotated bibliography The Content Analysis Guidebook Online, An Accompaniament to The Content Analysis Guidebook by Kimberley A. Neuendorf. The Association for Computers and the Humanities and the Literary and Linguistic Computing eximancer - Practical Text Mining and Concept Mapping Journal Practical Assessment, Research and Evaluation: some online articles Content Analysis News and Discussion mailing list archives some Resources related to content analysis and text analysis; updated quite recently: June 30, 2005;

That is because Ajax, which the Word Processor is based on, has to be "tweaked" for each web browser in order to work. First it is Mozilla Firefox, maybe next they will support Safari, Opera, IE 6.0, etc., but only after writing a modified version for those browsers and have the web site detect the browser type and load the correct Ajax script. I already had this discussion on the Microsoft Atlas story on the limits and compatability of Ajax and Javascripts.

As someone else noted, this is basically a Wordpad type replacement. No spell check, no grammar check, no advanced features that MS-Word users have relied on.

Corel Java Office was once in Beta testing, but Corel removed it. OpenOffice.org is written mostly in Java, but its Word Processing ability is a lot more advanced than AjaxWord.

If you want more than a Wordpad, and you don't mind downloading FOSS try AbiWord it can edit and write Word documents as well. It has been ported to Windows, Linux, OSX, etc. Unlike OO.org, it has a small footprint.

Here's someone who kept the old Corel Java Office. I remember being cautiously exited about this, but it turned out the computers of that time and the bandwith generaly available were a killer for this app (pun intended)...

Cheers...

FWIW, if you want to try it out, it's available here:

http://www.somis.dundee.ac.uk/pub/corelindex.htm

The past of web-based office suites...

Full Text Just incase

I am NOT an expert in throat cancer, but Dr. Hupp's homepage indicates that he is focusing on regulatory proteins associated with tumor formation, which has led to "the discovery of a novel stress-protein response in human oesophageal epithelium".

The "dipstick" test is probably one or more antibodies and an associated color reaction.

You could do it like a real caver (i.e. spelunker). Here are some links: Good summary of technique , Some relevant books, A cave surveying mailing list, Some software to reduce your raw data, Links of links to more of the above and cave equipment suppliers.

Don't you know the post-pc era happened 7 years ago? Isn't everyone running java thin client machines? Heck, I do all of my office work through my web browser using Corel Java Office.

I moved to the UK this year and switched from helping to create a startup telco to working at my wife's University. I've loved the switch to research and would have to think very hard about moving back. Some posters have already commented on the politics but I have to say that compared to a startup they are not bad at all. The pace of work is much better and incredibly more interesting.

• Bioformatics Dept.

I moved to the UK this year and switched from helping to create a startup telco to working at my wife's University. I've loved the switch to research and would have to think very hard about moving back. Some posters have already commented on the politics but I have to say that compared to a startup they are not bad at all. The pace of work is much better and incredibly more interesting.

• Bioformatics Dept.

"As you go from east to west the individuals change slightly, but can still interbreed (which is, more or less, the definition of what a species is)."

Not really. Ever hear of a liger?

It's a cross between a lion and a tiger. Two distinctly unique species can interbreed.

Horse + burro = jackass.
Severum (Heros Severus) + Red Devil (Amphilophus Labiatum) = Blood Parrot Fish.

There are several other examples of different species interbreeding.

Most commonly this happens with humans.

Caucasians, Mongoloids and Negroids interbreed more prolifically than any other group of species.

I know it isn't politically correct to say such things, but that's one of the main reasons I love science. It has no room nor desire for political correctness.

And before you oversensitive liberals MOD me into oblivion, know this: I'M BLACK

Knunov