That's the plot of lot of sci-fi stories, but on most of them the virus just affect all humans and get out of control.

But lets say that my Umbrella corporation wants to target a specific human population (e.g or people of color or with certain traits) and they have all the resources money can buy, all the best high security labs in space or deep sea whatever, all the best experts, and all the time to do tests and trials (all that people that disappear mysteriously must go somewhere right ?).

Could we develop a virus that target a specific human population and not all humans with current technology or within 1 or 2 centuries ahead from now ? What would be the most likely technique used by them to achieve that effect (e.g virus only active with certain conditions, or certain temperatures if you want to kill people in Africa but not in Greenland, a virus that infects everyone but only kills when certain chemicals are present, etc) ?

Perhaps this should be a new question, but would be possible to design this virus in a way that it won't mutate with time and extinct humanity?

It's okay to infect non-targets but they should only cause diseases in the target group and they better be 100% sure about this. Main targets would be Arab populations and people with special needs.

  • $\begingroup$ I think it depends entirely on which group you want to target. $\endgroup$
    – Erik
    Feb 1, 2016 at 16:06
  • $\begingroup$ I agree with Erik, precise targeting of a group is by definition group specific. $\endgroup$ Feb 1, 2016 at 16:10
  • $\begingroup$ You mean like lab-made sickle-cell disease – on steroids? $\endgroup$
    – Crissov
    Feb 1, 2016 at 16:33
  • 1
    $\begingroup$ Part of the problem with this question is the religion comment. Religion is not a biochemically selectable trait. So I would remove that, and then you should provide a scenario, including targets, and location. That would be an answerable question, as it stands, just to many options. $\endgroup$
    – James
    Feb 1, 2016 at 17:02
  • $\begingroup$ OK, I since I figured out this could be a delicate topic, I choose to made it a little broader but if the community thinks it'll make my question better I'll add it. @Crissov kinda that, but it shouldn't kill ( it's okay to infect only ) who they don't want to, their objective is not the extinction of human race, they are more like Hitler with steroids, so they better be sure it wont backfire when released. $\endgroup$
    – Freedo
    Feb 1, 2016 at 20:48

5 Answers 5


Caution: This is a delicate topic. Keep it civil in the comments, please.

This has been explored in other science fiction, such as the Doctor Who 2005 Christmas Invasion, in which the antagonist developed a virus that targeted humans with a specific blood type, resulting in 1/3 of the population being affected.

So that was a virus targeted by blood type, but depending on how you want to define the target group, the level of technology required could vary quite a bit.

Ethnic bioweapon

The most obvious way to "target" a population is selectively by ethnic group. This is unfortunately not a new idea. See ethnic bioweapon on Wikipedia for more information, but, for example, in the 1700s, while Europeans had relatively good immunity to smallpox, the Native Americans did not, which some people took advantage of.

As shameful as this (and other similar atrocities) are in our world history, the principle is unfortunately rather basic. Find a communicable disease that your target group has had little or no exposure to, and introduce it.

By region

Targeting a population by region would be more easily accomplished with weapons of mass destruction (WMDs). These include the usual suspects such as thermonuclear bombs, radiological agents, and chemical weapons. There is also overlap with bioweapons, but in this case the bioweapon would more or less have to target all humans.

By belief

Targeting people who, say, practice a certain religion, have certain political views, etc., would be much more difficult, because the determining factors are, at best, only partially genetic, and mostly environmental. See, for example, twin studies for general observations on nature vs. nurture, which you ought to be able to extrapolate to how difficult that would be for your targeted contagion.

  • 1
    $\begingroup$ Regarding targeting by religion: one could engineer a virus which is only deadly when the infected person consumes some kind of food or drink which is taboo according to the faith of the engineer. $\endgroup$
    – Philipp
    Feb 1, 2016 at 21:50
  • $\begingroup$ Or much simpler: release a short-living bioweapon in places of worship of the faith you want to target. But that would be kind of lame because any stock bioweapon would do in that case. $\endgroup$
    – Philipp
    Feb 1, 2016 at 21:56
  • $\begingroup$ Side note: The idea that white people deliberately infected the Indians with smallpox is probably an urban myth. Many historians have published articles saying that it is unsubstantiated. For example, quod.lib.umich.edu/p/plag/…. The Indians WERE decimated by smallpox, but there is little or no evidence that it was deliberate rather than an accident. $\endgroup$
    – Jay
    Feb 1, 2016 at 22:12
  • $\begingroup$ @Philipp RE any stock bioweapon: Or a good old-fashioned bomb. I suppose a bioweapon would have the advantage that the infected could spread the disease to the larger community. But unless the community you want to target is very insular, you'd be infecting many non-members at that point. $\endgroup$
    – Jay
    Feb 1, 2016 at 22:50

The best chance would be to have a virus that targets specific allel that are found primarily in the ethnic group you are targeting.

This wouldn't be fool proof, since many other might have it as well. It would take some research to find such a sequence, and then on top of that, you need to design a virus that NEEDS that particular gene sequence in order to multiply.

The 2nd part would be much more difficult, and with viruses no guarantees it wouldn't mutate.

Now, the one you suggest about chemicals being present, might work both as present or absent, say melanin. It could be attracted to it to kill off those with darker skin or repulsed by it to kill of those with lighter skin. And according to this, there are 6 genes that affect melanin so picking one or two of those might narrow down a target ethnic group.

EDT: Didn't see the hard science tag, I am unaware of having that kind of manipulation and design ability in viruses at this time. These are things that are theoretically possible.

  • $\begingroup$ Adding a +1 for the viral mutation mention. Viruses can, and will, mutate as they spread. They are incredibly... mutable in that regard (this is why you need a new flu shot every year). $\endgroup$ Feb 1, 2016 at 20:27

I am afraid "Arabs" are pretty much the worst possible target for genetically targeted bioweapon. First, as a side-note, the reason I have "Arabs" in quotation marks is that they are very diverse. They are really a collection of different ethnic groups that have common cultural and linguistic elements due to historical reasons. I am not an expert so I cannot say how much of a shared identity they have, but they are genetically diverse and not really distinct from their neighbours. So your bioweapon would only affect some Arabs and would probably affect lots of non-Arabs.

Looking deeper the problem becomes only worse. Middle East is the exact spot where humans migrated out of Africa. It is also the spot where cross-breeding with Neanderthals and Denisovans supposedly happened. And where humanity split and migrated East, North, or West. And later it was the hub of trade routes to West along the Mediterranean, East to India and South to Eastern Africa.

I will flat out tell you that a bioweapon that reliably kills most Arabs but is safe to Europeans, Indians, and Africans is not going to happen. Arabs are both too diverse and too connected to other ethnic groups.

People with special needs is even worse. They are special, which means they are not really a single group in any real sense. There are a multitude of reasons person could have special needs and you would need to target every reason separately. Even if you were willing to make the necessary investment, you'd probably develop the technology needed to cure them instead as a side-effect of the needed research. Which instead of making you into the most wanted terrorist in history would make you a great and very popular humanitarian making lots of money from his gene-therapy business. Looks like an easy choice to me.


I'm no biologist, but I'd think it is possible.

There are diseases that primarily or exclusively affect one ethnic group, like sickle-cell anemia (black people), or Tay-Sachs disease (Ashkenazi Jews). But those are genetic defects rather than diseases in the sense of something that can be spread by a bacteria or a virus. They are inherited, not caught.

I'm not aware of an existing bacteria or virus that only affects one ethnic group. But conceptually, I think it would be possible. If a virus attacked a particular gene, then only people with that gene would be vulnerable.

Or perhaps more practical, some ethnic groups may be more vulnerable to a particular disease because of genetic differences. It has been observed that black people are more likely to have heart attacks than white people, while white people are more likely to get cancer. I don't think the underlying causes for this are known, it's simply a statistical observation. It seems at least plausible that one ethnic group may have a natural immunity to a disease that another would not just by the luck of the draw of what genes got passed to them over the millennia.

My first thought on seeing your question was that it should be at least theoretically possible to target an ethnic group, but it would be almost impossible to target a religious or political group as these things are not biological.

But thinking further, it might actually be easier to target a religious, political, or social group. It seems likely that such a group has some set of behaviors they practice that are different from non-members. If so, then perhaps you can find or create a disease that is transmitted in ways that only affect people with such a practice.

To take an obvious example, Christianity teaches that sex should be limited to monogamous marriage. So practicing Christians should be largely immune to sexually transmitted diseases, no matter how much they spread among those around them. Or, Muslims and Jews do not eat pork. So a disease that is transmitted through infected pork would not affect them. Etc. Of course this only works to the extent that the people actually faithfully practice the teachings of their religion. If you tried to wipe out the Christian population while sparing Muslims by deliberately infecting pigs, any Muslim who sneaks a pork chop when no one is looking could become a victim too.

Indeed, there are lots of conspiracy theories that go around about how the government or some other sinister organization deliberately created a disease or contaminated some product to wipe out a group they dislike. AIDS was invented to kill homosexuals; the CIA contaminated crack cocaine to kill black people; right-wingers poisoned Starbucks lattes to kill white liberals; etc. (Ok, I made that last one up.) I'm not saying I put any stock in these sort of theories, but the fact that they are even remotely plausible shows that some groups are more likely to do X than others. If someone suggested that a conspiracy group was deliberately poisoning beer to kill French people, we'd say, "huh?" But say they're deliberately poisoning beer to kill Germans, and at least it makes sense.

I hope you're working on a fiction story and not a plot to wipe out some group you dislike.


Since we're talking about fiction (we are talking about fiction here, right?), we, as the creators of the fiction, can make decisions about what is possible, and what has been achieved. "A key scientist has discovered a specific type of virus that..."

Actually achieving that will be far more difficult. Your chances might be only 3 in 10,000 of having something work. But, for the sake of interesting fiction, you could say that you've achieved that 3 in 10,000.

Let's look at this in detail:

virus only active with certain conditions,

Yes, although some conditions will be more feasible/believable than other conditions. Conditions revolving around the biology of the virus, and for the virus's interactions with it's environment (possibly involving the biology of the human) may be effective.

You may also want the virus to be able to infect everybody, but to have benign effects for a certain group of people.

certain temperatures if you want to kill people in Africa but not in Greenland,

temperatures: yes.
People in specific geographical coordinates (certain number ranges in the longitude and latitude scales), based on current political boundaries: no.
People who are experiencing certain environmental conditions that tend to be unique to certain geographical coordinate, like dry environments (yes), or environments with less healthy green grass (more challenging).

a virus that infects everyone but only kills when certain chemicals are present, etc) ?

In inactive virus that is then triggered: yes. This may be harder to implement, but could be feasible.

design this virus in a way that it won't mutate with time and extinct humanity?

The virus could kill itself off in 80 years or 160 years. More challenging, but could be believable.

Won't mutate: Don't count on this. Mutations are challenging to handle, and there are billions of people so there are lots of changes of mutations. Completing preventing mutation would be an aspect that is unlikely to be easy for a designing scientist to control.

Main targets would be Arab populations

What do you mean by Arab?

  • People who are physically located in the Middle East: possibly. Geography was mentioned above.
  • People who pray to Allah and do not recognize Jesus Christ as the savior of the world: no. A virus is unlikely to care about what a person's beliefs are.
  • People who have names like Mohamed Abdullah, Anwar, Ali, Bassim, Hadid, Hashim, Muhammad, Sadid, etc.: No. Why would a virus care whether the name reminds westerners of the Middle East?
  • People with somewhat darker skin than caucasians, but lighter skin than black Africans: Maybe. Strong doses of Melatonin might be an anti-body (helping the blacks to survive). Maybe whatever affects eye color or hair color might help those green-eyed blondes out there.
  • Males who wear turbans: maybe, if the lack of sunlight on the scalp had some affect. Could also affect baseball players who wear caps a lot.
  • Males who wear turbans: definitely, if turbans are made of some specific material, and the viruses really like close proximity to that material. Could also affect anyone else who might use that material, so affecting different sub-cultures could result in collateral damage.
  • Females who cover their faces: possibly, if a lack of exposed air/sunlight caused swelling, possibly affecting breathing.
  • People who don't eat pork: possibly. Pork could somehow provide a nutrient that lets the Pork Eaters survive. Expect to wipe out Muslims who follow Islamic diets that consist of only Halal meat. Expect collateral damage of Jews and Seventh Day Adventists (and probably some others, too).
  • People who might be living in Canada, who had a grandfather who lived in Saudi Arabia: That would be much more challenging. The virus isn't likely to care what nation your grandfather lived in. However, the virus might care about some genetic trait that is common to certain people. Then, even the move to Canada, and the separation of multiple generations, and possibly even conversion away from the religion, might not be sufficient to permit the person to survive. (Even if you had an anti-Arab slant, what would the motivation be to target such a person anyway?)

people with special needs.

Yes, you can target people with special needs. Possibly. Depending on what their needs are.

It's okay to infect non-targets but they should only cause diseases in the target group and they better be 100% sure about this.

I'm not sure that I understand the difference between "infect" and "cause diseases". Maybe you meant that it can "infect" (spread), but not cause symptoms/problems? If so, using one method to spread (to many people) and another set of characteristics to determine who gets problems, is potentially believable. Spreading and causing problems is not necessarily the same thing.

In fact, statistically, you probably have an incurable disease, or two, yourself: offhand I'm not finding the name of either, but there are a couple of diseases that are incurable, and spread through contact with human skin, including mothers affecting pre-born children, and I think the infection rates might be something like 84% of humanity for one and 79% of humanity for the other, so chances are very high that you do have one or both of them. In fact, for some time some medical experts contemplated wondered whether this was just part of the definition of humanity. However, there have been some people located without the disease, and it only does harm (not good), and spreading/infecting has been identified. If you've never heard about these, it's because healthy immune systems keep the effect of these diseases in check. So they never cause fatalities unless the person already has some other problem (being elderly or having some other disease/condition) that prevents the immune system from being in check. So these diseases don't ever cause death by themselves. (I'd be happy to update this answer if/when I come across their names again.) In fiction, you could have something that is similarly widespread, but has different effects.

Expect that any virus that will take out (kill) millions of targets will probably also take out some non-targets. Just how evil of a warlord do you want to allow this fictional character to be?

Disclaimer: I am not advocating trying to kill off all of the characteristics that are mentioned above. In fact, there is some strong evidence that I fit into multiple categories that were mentioned above. (So that is some personal incentive I have for such a weapon to not be actually deployed.) I'm only discussing from the perspective of helping fiction to be more scientifically sound.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .