Wikipedia:Reference desk/Archives/Science/2012 September 2

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

September 2

Non-fatal viruses

Are there any viruses that are not potentially fatal to humans first-hand? One that I can think of is HIV (which can allow other diseases to cause death), but are there others? Thanks. 64.229.153.184 (talk) 00:17, 2 September 2012 (UTC)

There are millions of viruses which don't use humans as vectors at all, i.e. the Tobacco mosaic virus doesn't even infect animals. Bacteriophages are a class of viruses that infect bacteria. --Jayron32 00:28, 2 September 2012 (UTC)

In before this gets removed as a request for medical advice. What Jayron said, basically. I'm not even sure that a significant portion of the viruses in existence are potentially fatal to humans. That would seem to be rather oddly genus-specific, and I think most viruses tend to use a rather small number of animal vectors. As I understand it, the principle of natural selection would also make killing the majority of your hosts something of a genetic/evolutionary dead end, wouldn't it? Evanh2008 ^{(talk|contribs)} 00:34, 2 September 2012 (UTC)

It's not intended to be for medical advice, just out of curiosity. 64.229.153.184 (talk) 00:52, 2 September 2012 (UTC)

Yes, it is not evolutionarily stable for a virus to kill most of its hosts (or at least to kill them too quickly). Among viruses that infect humans (which I think is what the question is intended to focus on), I believe there are many that are rarely lethal (I do not have a handy reference, a little help here?). As for references, the OP may be interested in Virulence#Evolution, and optimal virulence. Both of these articles describe the factors that promote and inhibit lethality in pathogens. Lastly, this is a perfectly legitimate (and interesting) question. It is in no way a request for medical advice, see User:Kainaw/Kainaw's_criterion SemanticMantis (talk) 01:52, 2 September 2012 (UTC)

Are you asking for viruses that simply can't kill humans, viruses that infect humans but don't kill, or viruses that make humans more likely to die but can't kill on their own? Someguy1221 (talk) 01:46, 2 September 2012 (UTC)

Infect but don't kill, sorry for not being clearer. 64.229.153.184 (talk) 01:55, 2 September 2012 (UTC)

A couple of examples: GB virus C and Transfusion transmitted virus. -- Scray (talk) 02:08, 2 September 2012 (UTC)

We all get influenza from time to time. Only in rare cases do flue viruses kill. Various forms of the herpes virus either produces no symptoms, just a rash when the human is under stress, or just a mild rash. Wickwack124.182.34.199 (talk) 02:14, 2 September 2012 (UTC)

That does not answer the original question. -- Scray (talk) 03:23, 2 September 2012 (UTC)

It depends on what the OP meant by "first hand" I suppose. Then he clarified it by saying "infect but don't kill". One could ask "is there any food known that doesn't kill?" Strickly speaking, it is possible I could choke on a piece of apple, therefore apple can kill. I know a fried who is alergic to nuts. She inadvertently ate a nut in a cake at a party once, and had to have an emergency traecheotomy. So, do we say apples and nuts are lethal? That would not be of any value. My friend's root problem is not nuts, it's a genetic mishap. Similarly, flu viruses should not be considered lethal, though in rare cases they will if there is some other factor. I doubt that significant numbers of folk die from even the worst of the herpes types. This quite different to ebola virus, which will kill you regardless of other factors. So, yes, the OP's question has been answered. — Preceding unsigned comment added by 124.182.13.142 (talk) 04:30, 2 September 2012 (UTC)

Most viruses kill some, usually small portion of people who get them, such as chicken pox. Flu kills 5-50,000 a year in the US, not all that rare. http://www.cdc.gov/flu/about/disease/us_flu-related_deaths.htm Chicken pox around 100. Any infection can kill if you have a weakened immune system or a particularly virulent mutant. See Viral disease for a start at their variety. 02:37, 2 September 2012 (UTC)

No, that is incorrect. As I noted above, GB virus C and Transfusion transmitted virus are non-virulent. There are zero reports that they've killed. -- Scray (talk) 03:22, 2 September 2012 (UTC)

Are you telling me, Scray, that you are unfamiliar with the meaning of the word "most"? μηδείς (talk) 15:49, 2 September 2012 (UTC)

I should have been more specific in stating my disagreement: I was referring to your statement that "any infection can kill..." - because the viruses I cited are highly prevalent, many episodes of infection in immune-compromised hosts have been studied, and no deaths have been observed (citations in the respective articles, which I've already linked). I'll also point out that the original question was about viruses that are "not potentially fatal" - hence your answer overall was not responsive to the initial query (which implicitly acknowledged that most viruses can be lethal). -- Scray (talk) 21:51, 2 September 2012 (UTC)

Okay, that makes quite a bit more sense. It is true, however, that any infection that potentially causes an immune response can kill, although "infections" that fly under the radar, and neither evoke a response nor incapacitate their hosts obviously won't. But I am not sure that they would then be called infections. I was taking Tammy's response below as a great example. Warts don't usually kill. But papilloma virus infections of the lungs in AIDS patients do kill, and the virus does induce various cancers. (Oh, a point which I see on further reading you have explicitly mentioned, minus rectal cancer.) μηδείς (talk) 22:34, 2 September 2012 (UTC)

Warts are transmitted by a virus. --TammyMoet (talk) 08:33, 2 September 2012 (UTC)

True, but with respect to the original question, the viruses that cause warts can be lethal (e.g. PMID 10197157, PMID 21242344) to people with reduced immunity. More importantly, HPVs causes cancers of the cervix and the head and neck - also relevant: PMID 19646562. -- Scray (talk) 22:03, 2 September 2012 (UTC)

Adeno-associated virus "is not currently known to cause disease and consequently the virus causes a very mild immune response." Not sure our immune response really is a consequence of our knowledge, but that's what the article says. Consider also the article Viral vector, although this is a bit cheaty since those viruses have typically been engineered to be harmless.  Card Zero  (talk) 14:53, 2 September 2012 (UTC)

Some 95% of the adult population is infected with Epstein-Barr Virus yet it is rarely fatal unless it leads to for example Burkitt's lymphoma or HIV-associated lymphomas. Wolfgangus (talk) 22:22, 2 September 2012 (UTC)

what is a linear dielecric material

explain. --150.203.114.14 (talk) 04:23, 2 September 2012 (UTC)

Well, linear, in this sense, means an output variable varies proportionately with an input variable, therefore, if the input doubles, so does the output (although it could be slightly more complicated if the line on the graph is offset from the origin). However, I'm not sure what the input and output variables are, in the case of a dielectric. StuRat (talk) 05:32, 2 September 2012 (UTC)

A dielectric is a material that can maintain charge seperation; that is if I apply an external charge to it, the charges in the material shift around; applied negative charge will push negatives away, leaving positive in its place. What makes something a dielectric is, if the external charge is removed, the charge seperation stays for a time. Ever rub a balloon on your hair and stick it to the wall? That's the dielectric effect. I'm not exactly sure what property is linear, but my guess is that one could have a dielectric whose charge response is linear with respect to the input charge; double the input charge creates double the charge storage on the dielectric. --Jayron32 05:56, 2 September 2012 (UTC)

Jayron has given a physicist's answer. Here's a perhaps more useful and complete engineer's answer:-

What makes a dielectric a dielectric is that it displays a dielectric constant, normally termed permitivity. Permitivity ε in a linear dielectric is given in farads per meter by D/E where E is the applied electric potential across the dielectric, volts per meter, and D is the resulting electric field (flux), ie charge , Columbs per square meter. This is analogous to ferromagnetic materials, which have a permeability, in henries per meter, μ = B/H. Just as real magnetic materials are non-linear, that is permeability varies with the strength of the applied field, and also displays hysteresis, a real dielectric's permitivity varies with the applied volts per meter, and displays hysteresis - that is, you get a different permitivity at any given voltage when increasing it, to what you get when passing thru that same voltage while decreasing it.

A perfect vacuum has a permitivity of 10⁷/(4 π c²) F/m where c is the vacuum speed of light. All real substances have a permitivity above the vacuum value.

Just as a linear magnetic material allows the construction of an inductance having no electrical energy loss, a linear dielectric allows the construction of a capacitor with zero electric energy loss. It being a real (imperfect) world, you can't have either. However in practical electrical calculations, you can generally get a useful result with minimum effort by assuming they are linear, and then if necessary applying simple corrections to allow for the energy loss.

Keit60.230.231.168 (talk) 08:34, 2 September 2012 (UTC)

How to know that a drug is safe?

I know that clinical trials test new drugs in rats or other animals and then in people, who are paid for that. But, Independent of the model, human or animal, you are using, you'll be always extrapolating the results and taking risk, won't you? So, how could we know whether a drug is safe for pregnant women or children? OsmanRF34 (talk) 13:37, 2 September 2012 (UTC)

Actually, until you test it and wait for a couple of years or so, you can't know with total certainty if a new drug is safe for you. There can always be side effects that haven't been noticed during the trials. Here in Sweden, we have children suffering from narcolepsy after a vaccination against the swine flu that was supposed to be safe. Lova Falk talk 14:31, 2 September 2012 (UTC)

Furthermore, clinical trials seldom are longer than at the most half a year. Some meds are taken daily for years and years, and long-term effects are assessed while patients take the meds. For instance SSRI. They were tested for a relatively short time, put on the market, and now we can see if they are safe also for long term use. Lova Falk talk 14:35, 2 September 2012 (UTC)

• This is really just a specific case of the problem of induction, which applies to all scientific research. Looie496 (talk) 17:04, 2 September 2012 (UTC)

In many cases we don't know whether a drug is 'safe' (whatever that means) for pregnant women, many drugs ([1] has some numbers) are untested in pregnant woman and the information sheet that comes with the drug warns of that. As mentioned in the earlier source, the Food and Drug Administration Amendments Act of 2007 has a number of amendments that relate to drug testing in children contionuing and extending from earlier legislation, [2] mentions only about 20% of drugs approved by the FDA were labelled for pediatric use before they began a pediatric program (from that source I think that was in 1997). Nil Einne (talk) 17:45, 2 September 2012 (UTC)

I can add to that that a lot of research is done on men, because you know, women have their changing hormones and periods and that could confuse results, and then we just happily generalize the results to women, not bothering about possible differences. Also, a lot of the times, no trials are done with children. Yet they do get the meds. Lova Falk talk 18:31, 2 September 2012 (UTC)

As the sources I linked to and text of my reply hinted at, if no trials were done on children then usually the drugs won't be approved for use pediatric use and therefore any such usage is off label in the US and I think many other countries with decent regulatory regimes (see e.g. [3] for NZ and Australia). Nil Einne (talk) 19:31, 2 September 2012 (UTC)

Apart from acne and being fat, most indications for these recently approved drugs would justify the risk I think. Ssscienccce (talk) 08:12, 3 September 2012 (UTC)

• Actually there's a bigger problem than this - you don't know if drugs have synergistic toxicity. For example, the melamine contamination of products was such a problem because safety tests of melamine by itself suggested it was practically safe. But combine it with a related compound, and it forms huge flakes by noncovalent associations vaguely reminiscent of the base pairing of DNA. These flakes clog up the kidney tubules and kill pets or people. No drug ever placed on the market has ever been scientifically tested in combination with every other drug, chemical, and natural bioactive substance with which a user ingest, and so none of them are truly known to be absolutely safe. Wnt (talk) 02:27, 5 September 2012 (UTC)

Sunrise progression

I took a series of sunrise pictures from an east-facing beach in Florida last spring, including this closeup of the sun just as it rose above the horizon. My question is this: what causes the wedge-shaped base between the sun's disk and the horizon? It's definitely not a camera artifact. Juliancolton (talk) 15:02, 2 September 2012 (UTC)

It's a mirage, more specifically a mirage of an astronomical object caused by the bending of light. -- MacAddct1984 ^{(talk • contribs)} 16:55, 2 September 2012 (UTC)

Notice as well how the sun is not round. That bending of the light is caused by refraction due to the atmospheric inhomogeneity. Dauto (talk) 21:06, 2 September 2012 (UTC)

More specifically an inferior mirage, and quite pretty. μηδείς (talk) 21:07, 2 September 2012 (UTC)

"Pocket" flashlight that recharged from sun-light or room-light amout of photons?

automatically light-recharged flashlight, that lesser the natural need to replace a battery or that could even cancel it. just like solar calculators work. does anything like it or near it even exists? thanks. 109.64.151.131 (talk) 21:26, 2 September 2012 (UTC)

Sure. Google "Solar flashlight" You can buy one for under US$ 20.00 Dauto (talk) 22:36, 2 September 2012 (UTC)

Link? How long is the charge? μηδείς (talk) 23:48, 2 September 2012 (UTC)

Well, there are those night light things you put on the sides of your front walk, but they aren't very impressive. After absorbing sunlight all day on that tiny solar panel, they give a weak light for a few hours, at best. To get a bright light that lasts all night, you would need a huge solar panel. There are inefficiencies at every step that make this the case. When they are all considered together, you might be lucky to get 1% of the light that hits the solar panel back in light from the flashlight. StuRat (talk) 03:27, 3 September 2012 (UTC)

There are flashlights that use a capacitor and solenoid/magnet to charge. All you have to do is shake them for a minute or two and they will light up for a few minutes. NightStar Rechargeable Flashlight is one example 83.70.170.48 (talk) 10:36, 6 September 2012 (UTC)

What sort of skin diseases are there that look like melanomas?

I know there are cancers and other diseases of skin vascularization that are red-brown and have irregular outlines, but haven't been able to find anything on line by looking at our article for melanoma or for diseases that look like melanoma. I am aware of the various sarcomas. Are there any other suggestions? Thanks. μηδείς (talk) 23:47, 2 September 2012 (UTC)

I'm sure you'll spend many happy minutes looking at this. --TammyMoet (talk) 07:51, 3 September 2012 (UTC)

OH, my. Thanks. μηδείς (talk) 19:47, 3 September 2012 (UTC)

Resolved