I know that people say this but I have questions:
1. Are there concrete examples where drugs have been discovered from protein folding structures? What are the biggest ones?
2. Is there machinery that already exists to take in 3D protein structures and create drugs? or is this yet another issue?
3. How does folding of a protein in the current state impacting the use of the protein when it is used? Presumably these proteins are similar to polymers where they are not super rigid in all environments, how does the environment effect the protein folding?
Two of the first, back in 1999 [1], were Zanamivir and Oseltamivir (Tamiflu). Influenza neuraminidase inhibitors. Researchers examined a structure of neuraminidase co-crystallized with its substrate and designed a sialic acid paralog that was designed to bind with residues that are more conserved across different known sequences of influenza.
I found a recent review with some others listed here [2]. It has a nice overview of the process too!
Forgot to answer your other questions. I'm not up to date on the structure-based drug design workflow but back when I did similar work (5 years ago) there definitely were rudimentary systems for generating molecules and docking them. It may have improved significantly since then. But I would probably characterize it as a problem it itself for sure.
Your other question is a VERY good one. Proteins usually fold into whatever may be most favorable based on the sequence, and it mostly stays consistent once it does. However, they are very flexible and structures solved by EM or x-ray crystallography are like a photograph of bird flapping its wings: you will see the wings in a position, and if you happen to have a few birds in the photograph, you might get a sense of where those wings can move to, but it's never going to be perfect. But like wings, proteins usually still have a limited amount of movement. There are other types that are much harder to understand that have less structure, but globular proteins that bind to drugs like this are usually pretty well-predicted by the snapshots we can get.
For docking: Autodock Vina (from Scripps) is the most frequently cited docking software in the biomedical research literature. It's open source.
Researchers use docking software to run libraries of existing drugs as well as design never-seen-before drugs out of the enzyme protein's active-site pocket.
These operations have been performed extensively this year (by research groups all over the world) on covid's main protease enzyme as well as the spike-ACE2 interface, for example.
[followup] and so frankly, it's hard to imagine a world where drug discovery isn't enormously sped up by an automated protein-folding approach which docking software like Autodock Vina require to be run. I know that not all of the pharma industry agrees with this assertion however...:
https://twitter.com/michael_gilman/status/133375535280704307...
My take: Since 0.1% of proteins whose amino acids have been sequenced have ever seen a crystal structure (i.e. the folded model) generated of them. an automated approach to 3D model generation
1) will have enormous implications on drug development, and 2) will most likely come from a new and very different generation of drug developers, who don't have a lot in common with the generation that produced the tweet pasted above.
My take is exactly the opposite: since 3D structures of proteins alone are almost never the bottleneck in drug discovery, this won't actually change anything. Knowing how the drug is going to bind, and knowing how it'll behave in vivo, are not something you can predict from deep sequencing data.
Some proteins require help to fold properly (because there are more than one enrgetically favored conformations). The helping enzymes are called chaperons.
A known 3D structure for your target protein is very useful to improve molecules that bind to it, but we can't yet determine which molecules bind to a target without actually trying it experimentally. Of course there are methods to predict binding, but they not reliable enough and in the end the drug candidates are discovered by throwing a lot of molecules at a specific target or assay.
Once you have a candidate, it is very useful to determine the structure of the protein together with the drug candidate. There you can see how it binds, and can make some educated guesses on how to change the molecule to make it bind better, or to improve other aspects without making it bind worse.
Determing the protein fold from scratch without experimental data is impressive, but it doesn't have an immediate use for drug development. But a few steps further and it could certainly help if you can also predict which molecules bind to the protein structure.
I would strongly recommend the following blog post from Derek Lowe to put the importance of this into context for drug development:
The other point missing in most of these discussions is we already know how most drug targets fold, even if we don't know the exact structure at atomic detail. It's everything else about their structure, dynamics, and in vivo function that remains very difficult. The real promise in AlphaFold IMHO isn't that we can magically solve protein structures without experiments (most really interesting structures are beyond what it can do anyway), but the more general application of these AI methods to human health.
For the most part, people don't even try to make drugs for proteins that don't have structures, so that's one. As was mentioned in the anecdote, even as it exists AlphaFold can be an extremely powerful ally in structure elucidation in combination with lab methods. So this will help us increase the targetable list of proteins, especially the tricky ones that were harder to crystallize.
Once Alpha Fold or future programs get better with side chain modeling (not even for the entire protein just some parts), they will also allow complete computer based design of new antibodies against any target of choice (this is currently only possible through experiments and the technologies that allow this are all heavily patented and proprietary).
Variations of AlphaFold will also be significantly useful in research in general, potentially becoming fundamental enough that every project working with proteins might reach out to this tool like they reach out to to say mass spectrometry or flow cytometry.
From the article the pad produces 30 MW-hr/year/m2. This implies that you need 292533 m2 for this to produce more power than a 1 GW nuclear reactor.
For an idea of the scale of this water based power reactor that would produce this energy, I compare to the amount of space required for a large surface area reactor such as a CO2 chemical scrubber. These reactors have effective surface area of about 500sq meters of surface area per 1 meter3 of volume. This implies that you are looking at a reactor that has about 600 meters3 of volume. This is smaller than a typical olympic sized swimming pool, which is about 1700cm3.
I don't know if these comparisons are fair, but it implies that there are some questions about heat removal and how it might be as dense of a source as claimed.
Is he alive or dead? Cause of death? There are of course the radiothor incidents (where people actually died from eating uranium and thorium products, note this is selectively consumed for many years at concentrated levels, much higher than in rocks)[1]. Also the Baltimore Radium girls, who ingested concentrated radium to make glow in the dark watches.
Half-life is longer for uranium, but I would say they are just as radioactive, as they are in secular equilibrium and apart of the same decay chain [1]. They are just as "radioactive" in this configuration, as they go through the same number of decays. Their concentrations can change.
SI measures radiation in decay events per second. The rate is important because we mostly worry about decay product exposure during our lifetimes, or maybe how long a mass must be contained until it stops being an acute risk to our species, rather than the grand total of decay over astronomical time scales.
From 2015, scare mongering with little content other than vague statements like
"Before treatment, Westport's water tested up to four times state and federal limits. After treatment, it's safe for the children, teachers and staff to drink."
This does not point to a specific limit such as Maxiumum Contanminant Level (MCL) or Maximum Contaminant Limit Goal (MCLG). MCLG is a non-legally binding limit, which is as it states is a goal. MCLG for urainum contaminants is 0 ug/L. Any measurement of any drinking water will always have more uranium than this limit. The MCL is 30ug/L for uranium.
Additionally, the EPA has changed the way in which they have approached limits, previously they quoted something like the World Health Organization limits [3], which are average daily consumption limits. Average implies that some measurements could be high or low. WHO estimates natural concentration of uranium in drinking water is 0.08 ug/L (0.001 Bq/L), which is obviously above the EPAs MCLG.
WHO is a much better resource for information on these limits, and the EPA appears to just take their limits and do a divide by ten.
Why does low or negative yield bonds mean that you are going to be ok with govt bonds? This is exactly the problem, where bonds are no longer providing interest payments.
I agree that being more conservative is probably necessary, however I think other than specific investments and... burying your cash might be the "conservative" options. Bonds were those, and no longer are now.
> This is exactly the problem, where bonds are no longer providing interest payments.
If interest rates drop even more, the value of bonds go up.
Right now, a 1.68% 10-year bond looks like it sucks. But next year, a 1.68% 9-year bond will beat the pants off of a 1.3% 10-year.
You can sell a 1.68% 9-year bond for a lot more money when everyone else only has 1.3% 10-year bonds. If the 10-year drops to 1%, you'll make even more money. A falling interest rate market benefits those who buy bonds, especially if no one knows where the bottom is.
Thanks for the reply, I was confusing negative yield curves[1] with negative yield bonds (in the EU) [2].
It seems like if you already have high yield bonds, the value will continue to rise, as people exhaust other low yield options... however isn't this total value capped by the yield+face value? I don't really know how bond pricing works, but if you bottom out the yeild, or anticipated yeild... you should only be able to make the face value back or someone is buying negative yields.
EDIT: I _guess_ the bond reaches "maturity" in a much shorter time, which is perhaps your point. "1.68%" over 10 years is shit compared to "1.68%" over 1 year.
EDITEDIT: Also assumes you are not going to be eaten by inflation, which could push you into negative yeilds.
> It seems like if you already have high yield bonds, the value will continue to rise, as people exhaust other low yield options... however isn't this total value capped by the yield+face value? I don't really know how bond pricing works, but if you bottom out the yeild, or anticipated yeild... you should only be able to make the face value back or someone is buying negative yields.
1. Bond prices are primarily determined by auction.
2. If a big bank (and the US Fed is one of the biggest banks) decides to make a move, smaller banks, and the general market, will shift the prices of bonds.
> EDIT: I _guess_ the bond reaches "maturity" in a much shorter time, which is perhaps your point. "1.68%" over 10 years is shit compared to "1.68%" over 1 year.
No. Its 1.68% per year over 10 years. Bond pricing is standardized upon APY (its a "notational standard": bonds all have their own terms. But you can always math-out an effective APY given any bond structure). US Treasury Bonds physically have a coupon (every year, or maybe twice a year, they give a $$ amount), and a principle (at the end of the term, you get $$ back).
Anything less than 1-year only has principle (and is commonly called a "Bill"). So you get different APYs by shifting the price of the bill. Ex: You may buy a $1000 (principle) 1-year Bill for $980, effectively earning 2.04% APY in this hypothetical example.
In any case: the reason why a 1.68% 1-year is better than a 1.68% 10-year is because you only lock up the money for 1-year (in the case of the 1-year bond). So normally, a short-term bond gives a lower APY.
I know nothing of finance, but I have a normal liquid savings account that's paying 2.25%, apparently "permanently". Why would anyone buy a less-flexible product that pays less?
> normal liquid savings account that's paying 2.25%
Yes. That's what an inverted yield curve means. Liquid funds are "more expensive" than long-term funds. That's why things are inverted right now.
The long-term expectation (over the course of the next 10 years) is that savings accounts will drop. That's why people are willing to "only" be paid 1.6% for a 10-year, because its better to be paid 1.6% for 10 years... rather than 2.25% for this year (and then only 0.5% for the next 9 years).
In essence: the bankers are taking the opposite bet you're making. When the bankers are making a move, you probably should think about the future of money... bankers probably know more than you and I do.
EDIT:
> Why would anyone buy a less-flexible product that pays less?
Because they have a pessimistic view of the next 5 to 10 years. When big-money starts to make these pessimistic bets, its a recession indicator.
That high yield savings account is (likely) unavailable to their tax-advantaged account(s) such as 401k. In that case, their choices may be limited to stocks or bonds.
You can hold CDs in an IRA; currently Navy Federal is offering a 5-Year CD at 3.50% APY available for IRAs with no maximum purchase amount. [1]
Earlier this year I opened up an IRA with them and I put $100 in a 3.680% APY 40 months CD. I did this because they were matching the first $100 on new IRAs, so I deposited $100 and they deposited $100.
It's not actually permanent. Your bank almost certainly has the right to change it daily, and once their asset acquisition goals are met, they probably will.
Not knowing exactly how much you want explained; but...
The yield curve is inverting because buyers with serious money are buying medium-term cash instruments in defiance of naive valuation logic that the short-term cash instruments are more competitively priced.
This suggests that they see something in the near future, big enough that they are throwing the easily calculated "Net Present Value with usual assumptions" out the window when they make their purchasing decisions. Since bond buying and selling is usually done on a pure NPV basis this is a big deal and a good signal that it is time to avoid anything that might be risky until we find out what the big thing is.
Hence, buy government bonds as the single most conservative option. NPV might be partially irrelevant.
The bond interest payments are separate from the value of the bond. Which you can sell back at a greater premium and faster than even worrying about maturity.
This is the greatest bull market in bonds of all time and they are starting to behave like cheap deep in-the-money options contracts, which decline slightly in value over time due to theta (time value). Options are fun.
I’m not sure how I get your logic? I do think the US bond yields have a large discrepancy versus other developed market yields, but convexity greater increases as yields fall towards zero. So how is that behavior similar to a deep ITM option?
the only similarity is that a deep ITM option is an asset that decreases in value in one way while gaining in value in another way, due to the same forces.
a negative yielding bond - or a bond going towards negative yield - decreases in value one way while gaining in a value another way, the more it gains in value the deeper the negative yield goes.
TDP="Thermal design power"... (the amount of designed cooling power to get max cpu performance). ̶W̶h̶e̶r̶e̶ ̶i̶s̶ ̶t̶h̶e̶ ̶m̶y̶s̶t̶e̶r̶y̶?̶
Mystery might be also in the nuance of turbo vs base frequency draw. So even if you design for TDP as the "max thermal dissipation" you might get only base frequency performance.
On a side note people often confuse TDP with the max power draw of the processor. Intel does not specify total power draw probably because processor scaling is such an important consideration. If you have a system, the easiest way to figure out what max draw from the power source is to consult the power brick.
> Mystery might be also in the nuance of turbo vs base frequency draw. So even if you design for TDP as the "max thermal dissipation" you might get only base frequency performance.
That is really what is happening. If you have '4 core 3Ghz Base 4.5Ghz turbo 95W TDP' what that means is that the CPU will pull, by spec, a maximum of 95W with all cores running at the base frequency under full load. Turbo can exceed this power spec and maximum power draw under turbo is ill-defined.
> On a side note people often confuse TDP with the max power draw of the processor.
It's the same thing, any 'power' a CPU 'draws' is converted to heat, by definition.
Not every person that _starts_ and _makes_ a project successful grows _with_ the project. Being the person you were yesterday doesn't guarantee that fit the pattern tomorrow. I think that Linus should be the person to make this call. Perhaps Linux is where it is today because of Linus's low tolerance for BS, and he just notices that the components of the 'old' Linus are no longer acceptable/necessary going forward.
.svg){kind=link}