Science desk
< March 22	<< Feb \| March \| Apr >>	Current desk >

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

March 23[edit]

Wet Sciences Books That Would Be Good For A Mathematician[edit]

I was not sure how to phrase the title. I was a mathematician, but I left academia due to mental illness. I've always been fascinated by the "wet sciences", biology and medicine, in particular the immune system and viruses. While I can read a medical journal paper reasonably enough (research is reasearch), I don't really feel like I "get" these fields. With math, everything makes sense to me, it's abstract and clearly defined. Medical science, and related fields, on the other hand, seem like a pile of data that sort of fits together. I'm sure that is a defect in how I think and not the science, but I'd like to understand it better. So, I'm not looking for sources that cover the mathematics of these fields, but sources that would fit well with how a mathematician thinks - more abstract and deductive, if possible (physics has approaches like this, it feels like other sciences could). It doesn't need to be an advanced book, even intro stuff with this approach would be helpful. Thank you for the long read and any answers24.3.61.185 (talk) 11:35, 23 March 2020 (UTC)[reply]

During this interesting week, I had a little extra reading-time on hand, so I pulled an old book off my shelf - one that was gifted to me by my great friend, who is a clinician and an epidemiologist; and who, though trained as a physician in the United States, chooses to practice medicine overseas. The book is Complications: A Surgeon's Notes on an Imperfect Science. The presentation is not that of a textbook; but that may be the whole point - you have to adapt your way of thinking to fit the problems and solutions of a totally different discipline. Nimur (talk) 14:09, 23 March 2020 (UTC)[reply]

You might like to try "The Beautiful Cure: Harnessing Your Body’s Natural Defences".--Phil Holmes (talk) 14:52, 23 March 2020 (UTC)[reply]

If you consider neurology a wet science (after all it deals with wetware), I have found the writings of Oliver Sachs on neurological topics pleasantly perspicaceous. For sociobiology, perhaps enough of a biological topic to be on the wet side of life, try E. O. Wilson. I read Molecular Biology of the Gene by James Watson a long, long time ago; while I did not get everything, I thought it was an eye opener. I did not read his Molecular Biology of the Cell (not to be confused with the journal of that title), but on Amazon the current (7th) edition has 4.2 out of 5 stars, which is very high for a textbook. The 4th edition is freely available online, so you can get an idea before buying anything. Or perhaps it is already good enough; you're not studying to become a cell biologist. --Lambiam 15:49, 23 March 2020 (UTC)[reply]

Lots of biology makes more logical sense if you take an evolutionary perspective; that can help to place all the diversity of facts. I would recommend firstly Richard Dawkins The Selfish Gene. And, taking the message to medicine, try Nesse & Williams Evolution and Healing: The New Science of Darwinian Medicine, which I think is the same book as that with the title Why We Get Sick. Subsequent research has refined some of the details, but the principles hold true, and both books are beautifully and accessibly written. Jmchutchinson (talk) 20:30, 23 March 2020 (UTC)[reply]

We have an article (ok, redirect), mathematical biology. Added: In case this gives inspiration, Eric Lander, of human genome fame, started out as a research mathematician. 2601:648:8202:96B0:386A:A40C:EBB1:ACC0 (talk) 20:33, 23 March 2020 (UTC)[reply]

Speaking as a (neuro)biologist), I'm afraid trying to find mathematical rigour in something like biology is going to be futile effort. It not at all a defect in how you think, it's what the field is like. Our understanding is incomplete and basically empirical. There are no absolute certainties or proofs. I rather like that myself, but I can see how I can drive someone with a mathematical background mad. Having said that, Molecular Biology of the Cell as suggested above is a fantastic textbook, but probably too in depth for what you're after. Fgf10 (talk) 09:30, 24 March 2020 (UTC)[reply]

It's not a book, but I find this blog: [1] at the Journal of Public Health Management & Practice to be a good intersection of medicine and mathematics. --Jayron 32 14:32, 24 March 2020 (UTC)[reply]

Maybe you could also help straighten out the messed up state of statistics in health sciences. Look at John Ioannidis's publications, and Andrew Gelman's blog andrewgelman.com . 2601:648:8202:96B0:386A:A40C:EBB1:ACC0 (talk) 05:19, 25 March 2020 (UTC)[reply]

It occurs to me, you might like the book Probably Approximately Correct, by Leslie Valiant. We have an article Probably approximately correct learning which is only about the computerized aspect of this topic, but the book describes it as a mechanism of learning in nature, in humans and also in other organisms. That may help you get comfortable with thinking probabilistically instead of by pure logical deduction. Thinking, Fast and Slow also looks interesting. I haven't read it though. 2601:648:8202:96B0:386A:A40C:EBB1:ACC0 (talk) 19:59, 26 March 2020 (UTC)[reply]

Maybe you'd like the more mathy parts of chemistry. Pelirojopajaro (talk) 12:00, 28 March 2020 (UTC)[reply]

Viral lethality and evolution[edit]

From an evolutuionary view, is it disadvantageous for a virus to be highly lethal because the host's death prevents the virus' further spread (so that the natural selection selects those with low lethality to keep hosts alive)? Thanks. 212.180.235.46 (talk) 13:38, 23 March 2020 (UTC)[reply]

Isn't it amazing that with the power of the internet, we have near-immediate access to almost the whole of human knowledge - and equally amazing that for almost any interesting question, there is a high-degree of certainty that somebody in the scientific community has addressed it in great detail? In other words, if there is a meaningful question, there is a sort of ...pressure on a large number of professional researchers to try and answer it... and over the long run, one of them probably has answered it. These insights may help you to understand the nature of populations and selection pressure: there are no short-term guarantees, only long-term probabilities...

Here is the astonishingly relevant Evolutionary Insights into the Ecology of Coronaviruses (2007).

In case this is a little bit dense, let me try to summarize it in one sentence: on the scale of decades, virus evolution seems to be pretty random, and a lot less governed by selection-pressure - so while "survival of the fittest" makes for a very nice teachable sound-bite, the reality is much closer to "survival by a completely random bunch of 'em, with no specific identifiable benefit." "Natural selection" surely exists over the very long-term; but that ugly random-streak shows up in evolution and population-ecology way more often than they usually teach in high-school biology!

For further reading, the concept of a "reservoir population" of asymptomatic virus infestation is discussed here: Identifying reservoirs of infection: a conceptual and practical challenge (2002). It seems that this idea of a stable population of harmless viruses - "nonpathogenic... infections in reservoir hosts..." - well, this idea is conceptually pleasing but it's actually difficult to show in practice with any degree of convincing evidence.

Nimur (talk) 14:21, 23 March 2020 (UTC)[reply]

That's maybe a bit of a strong assertion, depending on what exactly is meant by the OP's question. Certainly much bacterial material is healthfully incorporated into our bodies, both at the level of mitochondria and gut flora. In the case of the gut flora, some of the little beasties involved are definitely infectious or harmful in other areas of the body but otherwise playing well with us. Matt Deres (talk) 17:48, 23 March 2020 (UTC)[reply]

It's worth remembering that evolution doesn't have an "end goal" in sight, and isn't trying to reach some future goal. It's a process acting only on current conditions, i.e. if I survive long enough to reproduce this generation, I reproduced during this generation. So, let's look at a highly lethal one, like Ebola. Ebola is transmitted best through contact with bodily fluids, so evolution may then favor something that increases that contact. A haemorrhagic fever works great for that, as you start bleeding out of every orifice, and then some, meaning there is a lot more opportunity for bodily fluid contact than would otherwise be the norm. So, it transmits to new hosts much more easily. A side effect of this is that it is highly lethal, since bleeding out of everywhere tends to kill you quickly. That means Ebola outbreaks tend to burn themselves out quickly, running out of hosts. Ebola will never be as globally successful as something with a far lower mortality, like influenza or SARS-CoV-2, but within its local population and region, it has a highly effective method of reproduction and transmission to new hosts. As long as it also has some sort of reservoir between outbreaks, be that zoonotic, or some sort of surface or buried material (think spores, like anthrax, but something with a virus), it will continue to exist and occasionally crop up. --OuroborosCobra (talk) 14:38, 24 March 2020 (UTC)[reply]

Selection works only on variants whose advantages or desadvantages are large. If they are not large they are indifferent and the variant frequency will drift. It is possible that the advantage in making the host very sick is large: so e.g. plague spreads best if the bacterial load of blood is large so that fleas have a larger chance of picking up some yersinia with a blood meal. That this large blood load will kill the patient is secondary, as long as large numbers of fleas are freshly infected before the patient dies. As Cobra says, an intermediate/reservoir host helps, but killing the patient ist really a desadvantage only if you kill them before they can reach you to the next victim. Ebola outbreaks ended quickly more because the symptoms were so serious that the patients were quickly insulated and less because the virus ran out of hosts, that is, they ran out of hosts because of the insulation. Without insulation measures any epidemic can destroy large populations, as did smallpox for Native Americans. 89.204.130.58 (talk) 16:50, 25 March 2020 (UTC) Marco PB[reply]

Yup. Bubonic Plague is actually a terrific example for this in another way as well. It kills its intermediate reservoir/host even more than it does us. It builds up a biofilm in the flea gut that eventually blocks up the gut completely, preventing it from digesting material. The flea, out of starvation, will try to feed again, which only results in regurgitation of bacteria into a new host, infecting them. The flea will eventually die from starvation. Bubonic plague literally kills its primary transmission vector, but as long as it only kills it after infecting at least one host, the bacteria will have successfully spread. Then, multiple uninfected fleas feed on the new host, and the cycle repeats, now able to spread to multiple new victims. --OuroborosCobra (talk) 15:22, 26 March 2020 (UTC)[reply]

See optimal virulence. --47.146.63.87 (talk) 19:05, 25 March 2020 (UTC)[reply]

One has to consider the evolution of pathogens and also the evolution of the host. As pointed out here, pathogens induce a fitness competition between the hosts without being able to eliminate pathogens. Count Iblis (talk) 12:47, 30 March 2020 (UTC)[reply]

Newbie audio panning[edit]

I have an MP3. The stereo is actually two mono channels, recorded via breakout adapter. I need to pan the two mono channels, 25-75 and 75-25. I'm on Linux. I'd prefer a command-line, or script solution. Thanks! — Preceding unsigned comment added by 182.156.108.31 (talk) 15:56, 23 March 2020 (UTC)[reply]

You can do that with ffmpeg but be ready to spend a month or two puzzling over the man page. The trac wiki (trac.ffmpeg.org) is very helpful in figuring out stuff like that. 2601:648:8202:96B0:386A:A40C:EBB1:ACC0 (talk) 20:50, 23 March 2020 (UTC)[reply]