The Chinese Room Argument, Understanding Math and the imposter syndrome

The Chinese Room Argument

 was first published in a 1980 article by American philosopher John Searle. He imagines himself alone in a room following a computer program for responding to Chinese characters slipped under the door. Searle understands nothing of Chinese, and yet, by following the program for manipulating symbols and numerals just as a computer does, he sends appropriate strings of Chinese characters back out under the door, and this leads those outside to mistakenly suppose there is a Chinese speaker in the room.

 

So it was with me and math as an undergraduate due to a history dating back to age 10.  I hit college being very good at manipulating symbols whose meaning I was never given to understand.  I grew up 45 miles from the nearest synagogue.  My fanatically religious grandfather thought it was better not to attend services at all than to drive up there on the Sabbath.  My father was a young lawyer building a practice, and couldn’t close his office on Friday.   So my he taught me how to read Hebrew letters and reproduce how they sound, so I could read from the Torah at my Bar Mitzvah (which I did comprehending nothing).  Since I’m musical, learning the cantillations under the letters wasn’t a problem.

 

I’ve always loved math and solving problems of the plug and chug variety was no problem.  I’d become adept years earlier at this type of thing thanks to my religiously rigid grandfather.   It was the imposter syndrome writ large.  I’ve never felt like this about organic chemistry and it made a good deal of intuitive sense the first time I ran into it.  For why have a look at — https://luysii.wordpress.com/2012/09/11/why-math-is-hard-for-me-and-organic-chemistry-is-easy/

 

If there is anything in math full of arcane symbols calling for lots of mechanical manipulation, it is the differential geometry and tensors needed to understand General relativity.   So I’ve plowed through a lot of it, but still don’t see what’s really going on.

 

Enter Tristan Needham’s book “Visual Differential Geometry and Forms”.  I’ve written about it several times
and Here — https://luysii.wordpress.com/2022/03/07/visual-differential-geometry-and-forms-q-take-3/

 

If you’ve studied any math, his approach will take getting used to as it’s purely visual and very UNalgebraic.  But what is curvature but a geometric concept.

 

So at present I’m about 80 pages away from completing Needham’s discussion of general relativity.  I now have an intuitive understanding of curvature, torsion, holonomy, geodesics and the Gauss map that I never had before.   It is very slow going, but very clear.  Hopefully I’ll make it to p. 333.  Wish me luck.

The silence is deafening

3 weeks ago I published a post about a paper that I thought would be a real bombshell, in effect contradicting a paper in a prestigious journal, and strongly arguing from real data that the pandemic virus could have been made in a lab, quite possibly Wuhan.  .

Absolutely nothing has happened. No letters to PNAS (the source of the article) to Cell (the source of the criticized study).  With a question of this magnitude and importance  you’d think Nature or Science would weigh in about it.  The origin of the pandemic virus is certainly they’ve covered extensively.

So I’m going to send this to all concerned and see if I get any feedback.

Here is the original post.

Evidence that the pandemic virus was made in a lab

 

Everyone knows that the Chinese have been less than forthcoming about the origin of the pandemic virus (SARS-CoV-2).  An article in the current Proceedings of the National Academy of Sciences — https://doi.org/10.1073/pnas.2202769119 arguesthat US data, which hasn’t been released, and some 290 pages of which has been redacted could shed a good deal of light on the subject (without any help from China).  One of the authors is an economist, but the other has serious biochemical chops — https://www.pharmacology.cuimc.columbia.edu/profile/neil-l-harrison-phd.

Basically a variety of US institutions (see the paper — it’s freely available) have been working with the lab at Wuhan for years modifying the virus, long before the pandemic.  The paper names the names etc. etc. and is quite detailed, but I want to explain the evidence that the virus could have been produced (by human modification) at the Wuhan lab.  It has to do with a site in a viral protein which says ‘cut here’.

Here is more background than many readers will need, but the virus has affected us all and I want to make it accessible to as many as possible.

Proteins are linear strings of amino acids, just as this post is a linear sequence of letters, spaces and punctuation.

We have fewer amino acids (20 to be exact) than letters  and to save space each one has a one letter abbreviation (A for alanine V for valine, etc. etc.).  The spike protein (the SARS-CoV-2 protein binding to the receptor  for it on our cells) is quite long (1,273 amino acids all in a row).

Our genome codes for 588  proteins (called proteases) whose job it is to cut up other proteins. Obviously, it would be a disaster if they worked indiscriminately.  So each cuts at a particular sequence of amino acids. Think of the protease as a key and the sequence as a lock.  One protease called furin cuts in the middle of an 8 amino acid sequence RRAR’SVAS (R stands for aRginine and S for Serine).  This is called the furin cleavage site (FCS)

A paper (The origins of SARS-CoV-2: A critical review. Cell 184, 4848–4856 (2021) argued that the amino acid sequence of the FCS in SARS-CoV-2 is an unusual, nonstandard sequence for an FCS and that nobody in a laboratory would design such a novel FCS.  So, like many, I skimmed the paper and accepted its conclusions, as Cell is one of the premier molecular biology journals.

One final quote “The NIH has resisted the release of important evidence, such as the grant proposals and project reports of EHA, and has continued to redact materials released under FOIA, including a remarkable 290-page redaction in a recent FOIA release.”

Sounds like Watergate doesn’t it?

 

Watch this space

BMOR is a bad actor

RNA and proteins have long been known to interact, but classic molecular biology pretty much had proteins down as something that modified RNA function.   Not so for BMOR, a long nonCoding RNA (1,247 nucleotides) expressed in breast cancer cells metastatic to the brain.  BMOR binds to IRF3 (Interferon Regulatory factor 3) inhibiting its phosphorylation by TBK1 with subsequent movement to the nucleus where it stimulates interferon expression which then turns on hundreds of genes producing inflammation.  All this is described in Proc. Natl. Acad. Sci. vol. 119 e2200230119 ’22 —

May 26, 2022
119 (22) e2200230119
Not sure if it is behind a paywall.    Definitely worth a read because knocking down BMOR in breast cancer cells prevents them from spreading to the brain (probably  by using BMOR to turni off the brain’s immune response to them).  Even more interestingly, BMOR was found to be only substantially expressed in breast cancer metastasis to brain tissue versus breast cancer metastasis to nonbrain tissues.

 

 

Teleology as always raises its head.  What in the world is the normal function of BMOR?  It can’t be what it is doing in the animal model described here.  Why would a cell make something to help it kill the organism containing it?

 

Then of course, as is typical of all interesting research, larger questions are raised.  Are there other RNAs whose function is to modify protein function?  Remember that 75% of the genome is transcribed into RNA.  Most of this has been thought of as molecular chaff, like the turnings of a lathe.   Time pick up the chaff from the factory floor and give it a look.

Brilliant structural work on the Arp2/3 complex with actin filaments and why it makes me depressed

The Arp2/3 complex of 5 proteins forms side branches on existing actin filaments.  The following paper shows its beautiful structure along with movies.  Have a look — it’s open access. https://www.pnas.org/doi/10.1073/pnas.2202723119.

Why should it make me depressed? Because I could spend the next week studying all the ins and outs of the structure and how it works without looking at anything else.  Similar cryoEM studies of other multiprotein machines are coming out which will take similar amounts of time.  Understanding how single enzymes work is much simpler, although similarly elegant — see Cozzarelli’s early work on topoisomerase.

So I’m depressed because I’ll never understand them to the depth I understand enzymes, DNA, RNA etc. etc.

Also the complexity and elegance of these machines brings back my old worries about how they could possibly have arisen simply by chance with selection acting on them.  So I plan to republish a series of old posts about the improbability of our existence, and the possibility of a creator, which was enough to me get thrown off Nature Chemistry as a blogger.

Enough whining.

Here is why the Arp2/3 complex is interesting.  Actin filaments are long (1,000 – 20,000 Angstroms and thin (70 Angstroms).  It you want to move a cell forward by having them grow toward its leading edge, growing actin filaments would puncture the membrane like a bunch of needles, hence the need for side branches, making actin filaments a brush-like mesh which could push the membrane forward as it grows.

The Arp2/3 complex has a molecular mass of 225 kiloDaltons, or probably 2,250 amino acids or 16 thousand atoms.

Arp2 stands for actin related protein 2, something quite similar to the normal actin monomer so it can sneak into the filament. So can Arp3.  The other 5 proteins grab actin monomers and start them polymerizing as a branch.

But even this isn’t enough, as Arp2/3 is intrinsically inactive and multiple classes of nucleation promoting factors (NPFs) are needed to stimulate it.  One such NPF family is the WASP proteins (for Wiskott Aldrich Syndrome Protein) mutations of which cause the syndrome characterized by hereditary thrombocytopenia, eczema and frequent infections.

The paper’s pictures do not include WASP, just the 7 proteins of the complex snuggling up to an actin filament.

In the complex the Arps are in a twisted conformation, in which they resemble actin monomers rather than filamentous actin subunits which have a flattened conformation.  After activation arp2 and arp3 mimic the arrangement of two consecutive subunits along the short pitch helical axis of an actin filament and each arp transitions from a twisted (monomerLike) to a flattened (filamentLike) conformation.

So look at the pictures and the movies and enjoy the elegance of the work of the Blind Watchmaker (if such a thing exists).

If the right hand don’t get you, the left hand will

Do you know the source of the title?  I found it surprising.  Answer at the end.

Some cancer cells have elevated levels of an enzyme called PHosphoGlyceride DeHydrogenase (PHGDH, others have decreased levels.  Many cancers contain both types of cells.  Neither is good news.

Those cancers  with low levels of PHGDH  have slower growth.  That’s good news isn’t it?  No.  Such cells are more likely to metastasize.

Those with high levels of PHGDH are less likely to metastasize.  That’s good news isn’t it?  No. such cells grow faster.

So cancers with both types of cells are more aggressive.

Here’s how it works [ Nature vol. 605 pp. 617 – 617, 747 – 753 ’22 ].

PHGDH is on the pathway for synthesis of serine, an amino acid required for protein synthesis (like all of them).  So low levels of the enzyme result in less protein synthesis and less tumor growth.

So how is this bad?  PHGDH binds to another enzyme PFK (PhosphoFructoKinase) stabilizing it.  When PHGDH is low PFK enzyme levels are low, so the subsrate of PFK (fructose 6 phosphate) is diverted to making sialic acid, which modifies cell surface proteins making them more likely to migrate.

So blocking sialic acid synthesis reverses the effects of low PHGDH on cancer migration and metastasis — but it does potentiate cell proliferation.

You just can’t win

Things like this may explain other paradoxic and unexpected effects of enzyme blockade.

16 Tons by Tennessee Ernie Ford

Evidence that the pandemic virus was made in a lab

 

Everyone knows that the Chinese have been less than forthcoming about the origin of the pandemic virus (SARS-CoV-2).  An article in the current Proceedings of the National Academy of Sciences — https://doi.org/10.1073/pnas.2202769119 argues that US data, which hasn’t been released, and some 290 pages of which has been redacted could shed a good deal of light on the subject (without any help from China).  One of the authors is an economist, but the other has serious biochemical chops — https://www.pharmacology.cuimc.columbia.edu/profile/neil-l-harrison-phd.

Basically a variety of US institutions (see the paper — it’s freely available) have been working with the lab at Wuhan for years modifying the virus, long before the pandemic.  The paper names the names etc. etc. and is quite detailed, but I want to explain the evidence that the virus could have been produced (by human modification) at the Wuhan lab.  It has to do with a site in a viral protein which says ‘cut here’.

Here is more background than many readers will need, but the virus has affected us all and I want to make it accessible to as many as possible.

Proteins are linear strings of amino acids, just as this post is a linear sequence of letters, spaces and punctuation.

We have fewer amino acids (20 to be exact) than letters  and to save space each one has a one letter abbreviation (A for alanine V for valine, etc. etc.).  The spike protein (the SARS-CoV-2 protein binding to the receptor  for it on our cells) is quite long (1,273 amino acids all in a row).

Our genome codes for 588  proteins (called proteases) whose job it is to cut up other proteins. Obviously, it would be a disaster if they worked indiscriminately.  So each cuts at a particular sequence of amino acids. Think of the protease as a key and the sequence as a lock.  One protease called furin cuts in the middle of an 8 amino acid sequence RRAR’SVAS (R stands for aRginine and S for Serine).  This is called the furin cleavage site (FCS)

A paper (The origins of SARS-CoV-2: A critical review. Cell 184, 4848–4856 (2021) argued that the amino acid sequence of the FCS in SARS-CoV-2 is an unusual, nonstandard sequence for an FCS and that nobody in a laboratory would design such a novel FCS.  So, like many, I skimmed the paper and accepted its conclusions, as Cell is one of the premier molecular biology journals.

One final quote “The NIH has resisted the release of important evidence, such as the grant proposals and project reports of EHA, and has continued to redact materials released under FOIA, including a remarkable 290-page redaction in a recent FOIA release.”

Sounds like Watergate doesn’t it?

 

Watch this space

 

Another neuropharmacologic surprise.

Our genome contains 826 different genes for G Protein Coupled Receptors (GPCRs) which are targeted by at least 475 FDA approved drugs (Nature vol. 587 p. 553 ’20 ). Yet part of the fascination of reading the current literature is the surprises it brings.

Our basic understanding was that the GPCRs sit on the surface of the cell waiting for ligands outside the cell to bind to it, which produces a conformational change on the cytoplasmic side of the cell membrane, changing the way the GPCR binds to the G protein, triggering all sorts of effects inside the cell.

As far as I recall, we never thought that different GPCRs would bind to each other in the cell membrane, even though a single cell can express ‘up to’ 100 different GPCRs [ Mol. Pharm. vol. 88 pp. 181 – 187 ’15 ].  Neurons express GPCRs and some are thought to be involved in neuropathic pain

But that’s exactly what Proc. Natl. Acad. Sci. vol. 119 e2123511119  ’22  is saying.

First a few definitions, if you’re as rusty about them as I was

A cytokine is an extracellular protein or peptide  helping cells to communicate with each other.  A chemokine is an extracellular protein which attracts cells.

Our genome has over 50 chemokines.  Most are  proteins with about 70 amino acids. The are classified by where the cysteines lie in them.  We have 23 receptors for chemokines, 18 of which are GPCRs.   Binding is promiscuous — a given chemokine binds to multiple receptors, and a given receptor binds to multiple chemokines.

Clearly the chemokines and their receptors are intimately involved in inflammation which always involves cell migration.  Neurons express chemokine receptors GPCRs and some are thought to be involved in neuropathic pain.

We also know that the nervous system is involved in immune function, particularly inflammation.  One prominent neurotransmitter is norepinephrine, and a variety of receptors bind to it.  There are 3 alpha1 norepinephrine receptors (a, b and d), all of which are GPCRs.

What is so shocking is that alpha1 GPCRs bind to chemokine receptors (forming heteromers), and that this binding is required for chemokines to have any effect on cell migration.  Even more interesting is that binding of norepinephrine to the alpha1 component of the heteromer INHIBITs cell migration.

So how many of our 826 GPCRs bind to each other, and what effects do they have?

Reading the literature is like opening presents, you find new fascinating toys to play with, some of which may actually benefit humanity

 

A new way to look at ALS (thank God)

It’s always good when a new way to look at a basically untreatable disease comes along.  We’ll know soon if looking at filamin A will be useful for Alzheimer’s disease.  Here’s another:  something we’ve known about for years (polyphosphate) may be important in Amyotrophic Lateral Sclerosis (ALS).   I used riluzole for ALS, but never saw any benefit.  It may have slowed the decline, but riluzole never stopped disease progression.

It is stated that 10% of ALS is familial, but I think this is an overstatement.  Even so mutations in a variety of proteins(superoxide dismutase 1 (SOD1) TDP43, C9orf72) do cause ALS, and studying them has taught us a lot about ALS.  There is plenty of work to do.  In 2016 a mere 160 mutations in the 153 amino acids of SOD1 had been found, but we still don’t know how they cause ALS despite hundreds of papers on the subject.  The proteins have allowed us to make mouse models of ALS, by putting in one or the other of mutated SOD1, TDP43, C9orf72 in motor neurons (or in whole animals)

Some real gumshoe work led to polyphosphate [ Neuron vol. 110 pp. 1603 – 1605 ’22 ].  Obviously in ALS, the motor neurons die, but recent work has shown that motor neurons are killed by neighboring astrocytes (containing any of the 3 the mutant proteins), when they are cultured together.   Normal astrocytes don’t do this.

So a lot of hard work found that it was polyphosphate in the supernatant fluid that was the killer.

So what is polyphosphate?  It’s been known for years, and is found in ALL cells — bacterial, plant, animal.  It also produced abiotically in volcanic exudates and deep sea steam vents.  No one knows what it does, so it has been called a molecular fossil.  Again teleology should inform biologic research (but it doesn’t).  Polyphosphate must be doing something useful or it wouldn’t be present in all living cells.

Chemically, polyphosphate is a chain of HUNDREDS to THOUSANDS of phosphate residues linked by high energy phosphoanhydride bonds.

Like this —

HO – PO2 – OH  + HO -PO2 -OH –>  HO – PO2 – 0 – PO2 – OH + H20

— the – O – in the middle is the phosphoanhydride bond

The authors treated motor neurons in culture with polyphosphate and found that it killed 40% of them.  So what?  Schmidt’s law of pharmacology, says that enough of anything will do anything,  So they looked at the spinal cords of patients dying of ALS and found that polyphosphate levels were higher than in neurologically normal controls.

So it’s open season on polyphosphate. Finding out what it does in normal cells, finding out how it kills motor neurons, finding out if decreasing its levels will help ALS (it does in cultures of motor neurons but that’s a long way from a living patient).  It’s an entirely new angle on an awful disease, with no useful treatment.  There is simply an enormous amount of work to be done.

Watch this space.

 

 

The brain gets more complicated the closer you study it

When our tools were blunt, the brain looked a lot simpler than it does now.

Example #1:  Locus coeruleus (LC).  This is a tiny group of neurons deep in the brain.  It looks blue to the naked eye, if you’ve gone to medical school as I have and dissected a brain.  This is held to be due to neuromelanin produced by the neurotransmitter  it uses (norepinephrine).  Neurons using dopamine as a neurotransmitter also produce neuromelanin but it’s brown.  The two differ by just one oxygen atom.

The LC is tiny in primates, only 15,000 – 50,000 neurons depending on species and who you read..  The rat (where most experiments are done) has only 1,500 in a space 1000 microns (1 millimeter) x 200 x 500 microns.

So until now any attempt to stimulate the locus coeruleus with an electrode alerted the animal.

Why? Two reasons.

1. The LC is so small that any electrode stimulating it, stimulated every neuron.

2.  Because that tiny nucleus sends fibers all over the brain, releasing norepinephrine everywhere, and not just at synapses.  This is called volume neurotransmission. Most places on the axons of a LC neuron showing synaptic vesicles (where norepinephrine is found), don’t have a dendrite or any sort synaptic specialization next to them.  So the LC innervates the whole brain, in the same way that our brain innervates our muscles.

Stimulate the LC of a rat and the brain is flooded with norepinephrine and the animal wakes up.

Well that was the case until technology marched on and miniaturization of electrodes allowed us to record from a 10 – 20 neurons at a time in the LC when stimulation was applied.  Those responding to a given stimulus were called ensembles.  Of 285 single LC neurons studied in 15 rats, 115 participated in multiple ensembles, 149 participated only in a single ensemble and 21 didn’t participate in any.  Activity of different ensembles produced different brain states — not all were wide awake.  You can read all about this in Proc. Natl. Acad. Sci. vol. 119 e21116507119 ’22.

Volume neurotransmission is  important because the following neurotransmitters use it — dopamine, serotonin, acetyl choline and norepinephrine. Each has only a small number of cells using them as a transmitter.  The ramification of these neurons is incredible.

For instance, “human serotonergic neurons, which are estimated to extend axons for 350 meters”  [ Science vol. 366 3aaw9997 p. 4 ’19 ], so the fibers are everywhere in the brain.  I couldn’t find a statistic for axons of the locus coeruleus but those of neurons using acetyl choline as a neurotransmitter are estimated to have axons extending for 31 meters.

So now you see why massive release of any of the 4 neurotransmitters mentioned (norepinephrine, serotonin, dopamine, acetyl choline) would have profound effects on brain states.  The four are vitally involved in emotional state and psychiatric disease. The SSRIs treat depression, they prevent reuptake of released serotonin.  Cocaine has similar effects on dopamine.  The list goes on and on and on.

Maybe be we’ll be able to slice and dice these nuclei in the future to produce more subtle effects on brain function.

 

Example #2:  Dendritic diversity  — that’s for next time.  This post is long enough.

Cassava’s Cognition Maintenance Study may prove Simufilam works

The FDA will approve less than perfect therapies if there is nothing useful for a serious condition.  Consider the following from Proc. Natl. Acad. Sci. vol. 119 e2120512119 ’22

“KRAS is the most frequently mutated oncogene in human cancer, with mutations detected across many lineages, particularly in the pancreas, colon, and lungs. Among the most commonly activating KRAS mutations at codons 12, 13, and 61, G12C occurs in ∼13% of lung and 3% of colorectal carcinomas and at lower frequencies in other tumors.

“In locally advanced or metastatic non–small-cell lung cancer (NSCLC) patients with KRASG12C mutations who have received at least one prior systemic therapy”  treatment with sotorasib resulted in the following “objective response  in 37.1% of the patients, with a median duration of response was 11.1 months.”   This is hardly a cure, but nonetheless “This promising anticancer activity has resulted in accelerated approval from the US Food & Drug Administration”

Which brings me to the current CMS study from Cassava Sciences.  I’ll let them speak for themselves. https://finance.yahoo.com/news/cassava-sciences-reports-first-quarter-130000375.html

Cognition Maintenance Study (CMS) – on-going
In May 2021, we initiated a Cognition Maintenance Study (CMS). This is a double-blind, randomized, placebo-controlled study of simufilam in patients with mild-to-moderate Alzheimer’s disease. Study participants are randomized (1:1) to simufilam or placebo for six months. To enroll in the CMS, patients must have previously completed 12 months or more of open-label treatment with simufilam. The CMS is designed to evaluate simufilam’s effects on cognition and health outcomes in Alzheimer’s patients who continue with drug treatment versus patients who discontinue drug treatment. The target enrollment for the CMS is approximately 100 subjects. Over 75 subjects have been enrolled in the CMS and 35 have completed the study.”

Even though the open label study was not randomized, this one will be.

Only someone who has actually taken care of  patients would know the following.  People who are getting no benefit from a drug will soon stop taking it.  This was particularly true for my experience with Cognex for Alzheimer’s disease.

Which is exactly why the fact that 75 patients who’ve been on Simufilam have decided to continue on in the CMS study.  Presumably they feel they are getting some benefit.

There are two possible hookers to this

l. The patients are being paid to enter CMS

2. The original cohort was 200, not all of whom have finished the 1 year.  So we don’t know how many could have been in CMS but chose not to.

As I discussed in an earlier post, the most impressive thing (to me at least) was that at 9 months 5/50 had significant improvement in their cognition — here’s a link — https://luysii.wordpress.com/2021/08/25/cassava-sciences-9-month-data-is-probably-better-than-they-realize/.

The CMS study should give us an idea of how they fared at 1 year and  at 18 months.

If:

l. gains in cognition were maintained on Simufilam

2. gains in cognition were lost off Simufilam

FDA approval should follow quickly.

Results on the 75 will be available this year.   Also available this year will be 1 year results on all 200 entering the open label study.

There are two other double blind studies in progress which will provide  more definitive answers, but they are far from full and will take much longer to complete.  So stay tuned.