Hacker Newsnew | past | comments | ask | show | jobs | submit | Smileyferret's commentslogin

Interesting article, but in the full paper their key figure (Fig 2) shows their treatment group of n=3 mice completely responded to the bacterial treatment, but their methods say they treated n=5 mice? Could be an honest mistake but that’s a little concerning for data manipulation.

Also agree that using a PD-L1 mab feels like it’s for show especially considering the cancer model they’re using (Colon-26) was shown to be substantially less responsive to PD-L1 inhibitors…

Not the world’s best paper imo


Still the idea is beautiful. Since tumors are oxygen-deficient and suppress the immune response, anaerobic bacteria would proliferate there, and wreak havoc, while in the healthy parts of the organism they would be rapidly eliminated. Additionally, since the bacteria accumulate in the tumor, and the immune system has just responded to their invasion, T-cells will flock to the tumor, destroying what remains of it in due course.

As they say, "the fame of a mathematician is measured by the number of poor papers", because pioneering works are often awkward, treading completely unknown ground. Maybe the same applies to biology sometimes?


I'm untrained in this area but it sounds like similar ideas I heard of in the last decade where injected tumor with virus (polio?) that quickly kills the tumor but then is easily cured in the rest of the body since we have solutions to them. Is this a similar situation?


Figures 2 and 3 seem to be different experiments, with n=3 and n=5, respectively. Both showed 100% survival. Obviously very small sample sizes, but still promising.


Yes in figure 2 it's 3 mice, next figure 3 they also have 5 (panel e)


This is an abstract that hasn’t been peer-reviewed… based on prescription data for an over-the-counter medication. This will be horribly inaccurate because it will miss all the folks who just buy the medication on their own and never have it documented.

These TriNetX studies are usually garbage because they’re entirely dependent on how accurate/up-to-date the medical record is.


Yes, this should have been a pure dose-response study among people with any history of filled melatonin scripts.

The comparison between the US and UK probably leads to two issues - US users use way too much melatonin and swamp heart disease signal, while UK patients prescribed melatonin probably have significant sleep derangement (consider how much effort it takes to get prescribed something for sleep - you need to schedule an appointment, convince your doctor, go to the pharmacy, etc)


“The specific bias in the current context is the tendency for the public to think that homeless individuals will increase spending on temptation goods (alcohol, drugs, cigarettes) when given the cash transfer compared to people who are not homeless. This bias can favor the provision of paternalistic forms of aid over more agentic forms of aid, thus presenting a barrier to the cash transfer policy.”

Earlier in the paper:

“Our preregistered screening criteria were: … [history of homelessness less than 2 years and] nonsevere levels of substance use (DAST-10), alcohol use (AUDIT),and mental health symptoms Colorado Symptom Index (CSI) based on predefined thresholds. These screening criteria were used to reduce any potential risks of harm (e.g., overdose) from the cash transfer.”

… Are you kidding me? Talk about hypocrisy.


I guess they weren’t testing for whether the bias is true.

Though yeah, it’s a bit wishy washy to call something a bias and not real, then design your test to avoid challenging the notion.

A bias could still be present though. A phenomenon can exist but be overestimated or underestimated in degree due to bias.


People don't want to support cash transfers because they think that the beneficiaries will blow it on drugs. But they're wrong: if you eliminate anybody who would blow it on drugs, then that problem doesn't occur at all!


I think I've also heard of similar experiments where a significant number of the people who used drugs actually quit using drugs once they got other options.

I don't have a link, though. Just a vague memory of reading about this once.


Yeah, and we couldn’t disqualify people from UBI anyways, as that could mean banishment from an UBI-inflated economy. They might be unable to afford even the most basic necessities when the economy adjusts to everyone’s increased spending power.

So with UBI will come a certain amount of UBI spent on blow. And maybe that’s fine if overall humanity is better off. It doesn’t need to be perfect, just considerably better.


It’s much worse than wishy-washy. It’s purposefully deceitful.


I am rapidly losing all faith in "science" as it is currently practiced. Between the overwhelming bias for only producing "correct" findings and outright fraud for professional advancement, and the coverup that frequently occurs in both cases, I think the level to which good intentions are presumed and trust is vested is neither deserved nor constructive. We are at a point that to me seems familiar as in many other industries when we empowered financial and technical agencies to self-certify only leading to unbelievably brazen deception with disastrous outcomes.

It is a shame to see so much potential wasted while years tick by that we will never get back.


"Public thinks homeless will spend cash on drugs, this is false. Our trial excluded people with drug issues, and we found no evidence cash was spent on drugs".

Well done!


Are you saying they are wrong and the general public actually approves of unconditional cash transfers to homeless people /as long as they don’t have drug issues/?

I don’t personally see much support for giving cash to homeless people of any kind so I’m not seeing the hypocrisy in the statements here.

Do you support unconditional cash transfers to homeless people who are not drug users as defined by this study?


Gotta get an ethics board to approve your study, which might affect the efficacy of the study, which might be an ethical concern. Oh dear.

Interested to see this study redone, but also toss them a handful of narcan and a wearable AED.


Some interesting bits:

> We recently reported the development of a “universal” phage T4 vaccine design platform by Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) engineering that can rapidly generate multivalent vaccine candidates.

> Intriguingly, the T4-CoV-2 vaccine induced high levels of Spike-specific serum IgA antibodies (endpoint titers up to 62,500) when administered by either the i.m. [intramuscular] or the i.n. [intranasal] route. This is notable because IgA stimulation is not commonly observed in traditional vaccines…

> … significantly elevated percentages of CD4+ T cells producing IFN-γ were detected in the i.n. group (1%) in comparison to the i.m. group (0.55%) of vaccinated mice (P < 0.001 between i.n. and i.m.). These data indicated an enhanced Th1-mediated immunity induced by i.n. administration of the vaccine.

> … high titers of mucosal sIgA antibodies were elicited by i.n. vaccination (endpoint titers up to 12,500), in addition to high levels of systemic immune responses as described above. In contrast, i.m. immunization failed to produce sIgA, which is not unexpected.


This sounds like a very promising method of vaccination!!


I don’t disagree with you that much of medicine is hand-waving and voodoo magic, but arrhythmias due to scar tissue likely have a similar mechanism to other well-studied arrhythmias. Overall I think that cardiology has some of the more scientific explanations of its pathophysiology than other fields. Scarring simply interrupts smooth conduction of impulses throughout the myocardium and makes it more susceptible to arrhythmias. If you want to see a similar mechanism explained, check out this video:

https://youtu.be/j8pVU9snSH4


Thanks for the link.

I need to choose words better - rather than "dark arts" I probably should have said "tacit knowledge". I certainly didn't mean to imply that arrhythmias are not well studied, rather, my hope is that we can use AI methods to do better science.

side note: I know EEs who claim antenna design is a black art :-) It also is well studied and yet according to some, it appears magical at times.


https://en.wikipedia.org/wiki/Apparent_mineralocorticoid_exc...

We actually learned about this in med school last week, but I never thought it actually happened to people... Licorice contains glycyrrhetinic acid which inhibits the conversion of cortisol to cortisone. If you have too much cortisol floating around because of this, that cortisol starts activating receptors in your kidneys that are normally activated by aldosterone. This in turn increases your sodium resorption and potassium excretion, so you get high blood pressure and low potassium which can cause arrhythmias. It's called the Syndrome of Apparent Mineralocorticoid Excess and is a usually genetic problem with the enzyme that breaks down cortisol.


Can this imbalance occur transiently?

One time, I was awake for ~3 days straight. I ended up in the hospital (visual hallucinations) and the only thing remarkable was low potassium and high BP.

But I've had this long running hypothesis that body has trouble with cortisol regulation when I'm extremely fatigued. If I miss a night of sleep, the next night it is extremely hard for me to fall asleep (cortisol inhibits sleep) and I have that terrible 'stress sweat' odor.

Remeron will then knock me out... where as hypnotics will not. Remeron being interesting because it inhibits the production of cortisol.


I would suggest that the hallucinations were simply a symptom of your sleep deprivation. I've experienced the same after a similar amount of time awake coding.


I starting to suspect I have a variety of such genetic endowments- gout because I can't break down uric acid, high blood pressure that gets instantly elevated when I eat black licorice (which I loved as a kid), variety of allergies, Type II Diabetes (along with most of my sibs, regardless of their diets.)

Hopefully Dad's Parkinsons isn't genetic...


So it would me even more deadly if they were taking Prednisone?


Not necessarily. Believe it or not, dexamethasone (another corticosteroid) is actually a second-line treatment for this disease because it reduces endogenous cortisol production without binding too tightly to the mineralocorticoid receptor in the kidneys.


Would there be some way to use this to your advantage?


... if you had low blood pressure and potassium poisoning perhaps?


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: