Hacker News Reader

Comment by ksenzee | Hacker News Reader

The article quotes Denver's district attorney as saying that this familial DNA searching is especially good for "cold cases where the victims are women or children". Women or children. We're searching through databases of innocent people's DNA, finding potentially damaging false positives, all because we need to keep women and children safe. Has this guy been binge watching Mad Men?

View on HN

NOTE: the original submission of this was the fusion.net story. The mods changed it to a wired.com story, then later changed it back to the fusion.net story. It was the wired.com story when ksenzee read it. Only the wired.com story quotes the Denver DA.

Since the discussion of ksenzee's comment is the most popular branch of the discussion, I think there should be a link somewhere to the story ksenzee read, so here it is: http://www.wired.com/2015/10/familial-dna-evidence-turns-inn...

Didn't you know? Familiar DNA is only useful when the cold case victim is a woman or child. It's not actually any good if the victim was a man. That's just like, basic science, man.

You are right. The reason behind it is that these companies use mitochondrial DNA for their analysis, which is only passed down from mother to child. /s

But why do they need to analyze the victim's DNA - wouldn't they typically be looking for a match for the perpetrator of the crime?

I guess if the victim's DNA were on the offender... But then why not ask the victim for a sample? :/

Because the victim is dead and they can't find any family.

Because the victim is a John or Jane Doe (i.e., they don't know who it is).

Except they analyse Y-chromosomal DNA as well.

You're right that Ancestry.com is focused on mtDNA, but 23andMe actually uses a custom Illumina microarray with autosome, allosome, and mitochondrial targets. Here's a study where 23andMe used their (old version) chips to replicate a bunch of known genetic associations, most of which are autosomal: http://journals.plos.org/plosgenetics/article?id=10.1371/jou...

But they aren't interested in the victim's DNA so that is irrelevant. They are interested in other DNA found at the crime scene, i.e. belonging to the attacker or someone related to the attacker.

Yes. josu was being sarcastic.

What does /s mean?

It marks whatever precedes it as sarcastic.

I thought so, but wasn't 100% sure.

Thanks.

Showing the world that you pay constant attention to the safety and well being of women and children is a good way to assert how much of a good, real, honest, straight man you are to society. Along with loudly proclaiming "they should all be shot" whenever the word paedophile is mentioned.

Did you just go from creating a straw man to a "well, actually paedophiles are kind of redeemable" stance?

The gallantry and defense of children is because men that are in charge are predominantly straight and have children. Not everything is a charade.

A pedophile is not by definition a child molester/abuser. I wish they would get help and be redeemed from their abnormal urges.

to clarify this comment:

"pedophile" refers to someone with primary sexual attraction to children, whether or not they act on that attraction.

Terms like "predator", "molester", and "abuser" refer to someone who takes sexual actions against children.

There is overlap between the groups, but they're not identical. There are celibate pedophiles, and pedophiles who have ordinary adult relationships and keep themselves away from children. And there are predators who aren't specifically attracted to children, but instead are attracted to the power imbalance.

I get the feeling that some people want others to focus on the thought instead of the action, because their own actions are deplorable and they're just trying to convince everyone that their thoughts are pure.

Yes, that's been the case since the predominant alma mater of Hacker News became either Wikipedia State or Reddit University. It's no longer a place for discourse, but instead for people trying to get internet points based on falling back to general, unoffensive platitudes.

I think people draw the distinction because there's a huge difference between finding someone sexually attractive, and sexually assaulting them. (This applies to attraction to adults, not just children. "I find that person attractive" is very different from "I'm going to have sexual contact with that person without their consent".)

Paedophiles only enter popular conversation in response to one doing something unredeemable (acting on their urges). Of course I'd people to get treatment before someone is hurt.

…as oposed to the women who, if there were more of them in significant political positions, would not be predominantly straight and have kids and therefore would not be so defensive of children?

No, not opposed to those women, but the wording would be "family" instead of "women and children."

> Did you just go from creating a straw man to a "well, actually paedophiles are kind of redeemable" stance?

No, but you went to a straw man very quickly. In addition, it's clear you don't understand statistics.

First, DNA is often far from definitive. So, if the error rate is 1 in 30,000, you're very likely to get quite a few hits of innocent people when the perpetrator isn't in the database.

Second, people who use "women and children" or "pedophiles" as justifications can generally be considered to be engaging in dishonest behavior. Something that has a logical justification doesn't have to hide behind emotional appeals.

Third, "pedophile" or "sex offender" are terms you can get saddled with very easily in the US for a whole range of behaviors including things like giving a child a lost child a hug to comfort them (oops: now you're a pedophile) or taking a whizz in public (congrats: you're a sex offender).

Finally, even the worst and most guilty among us deserve to have their rights protected because those are OUR rights when WE get accused of something heinous of which we are not guilty.

Please explain to me how I don't "understand statistics." In no way was I addressing or defending the wholesale scanning or mining of data; I was simply addressing uhtred's statement.

Politics is largely the use of emotional appeal in a democracy -- you don't get elected on a stance of "well, if we understand the data."

>Third, "pedophile" or "sex offender" are terms you can get saddled with very easily in the US for a whole range of behaviors including things like giving a child a lost child a hug to comfort them (oops: now you're a pedophile) or taking a whizz in public (congrats: you're a sex offender).

Please, link me to a case in which a person giving a lost child a hug led to charges. I would earnestly like to read it because I think you're being clumsily dishonest.

>Finally, even the worst and most guilty among us deserve to have their rights protected because those are OUR rights when WE get accused of something heinous of which we are not guilty.

And now you're just using a truism as frosting to your argument. You're framing this in a "either you care about paedophiles or you don't care about OUR rights."

>Please, link me to a case in which a person giving a lost child a hug led to charges. I would earnestly like to read it because I think you're being clumsily dishonest.

Pedophile isn't a legal term. Men get "straddled with the term pedophile" for simply being in a playground.

No one gets charged with being a pedophile because being a pedophile is not illegal.

>Pedophile isn't a legal term. Men get "straddled with the term pedophile" for simply being in a playground.

I sat on a swing earlier this summer and I've never been accused of such things.

You created this "man hugs lost child, gets labelled a paedophile" scenario that's just ridiculous.

I never mentioned hugs.

But if you want examples of men being accused of being pedophiles for being on playgrounds, or similar things check out these:

The first 7 are playgrounds/parks, and the rest are other similar situations.

https://www.reddit.com/r/AskReddit/comments/xhwt8/i_just_saw...

https://www.reddit.com/r/MensRights/comments/2ayvaf/kicked_o...

https://www.reddit.com/r/AskMen/comments/24npnm/men_in_what_...

https://www.youtube.com/watch?v=QaV4saiEHkQ

https://www.reddit.com/r/AskReddit/comments/xhwt8/i_just_saw...

https://www.reddit.com/r/AskMen/comments/24npnm/men_in_what_...

https://www.reddit.com/r/AskReddit/comments/xhwt8/i_just_saw...

https://www.reddit.com/r/AskMen/comments/24npnm/men_in_what_...

https://www.reddit.com/r/AskReddit/comments/xhwt8/i_just_saw...

https://nakedsecurity.sophos.com/2015/05/11/creep-shamed-on-...

Youtube and reddit have a shaky relationship to reality.

"people on the internet aren't real"?

My own comments on the topic:

https://news.ycombinator.com/item?id=4133027

https://news.ycombinator.com/item?id=6003797

> cold cases

That is the terrifying part. Does anybody known how much data these databases are actually storing about someone's DNA, after processing? I ask because in the case of some forensic DNA matching techniques, the set of possible results is small enough to have birthday paradox problems.

Matching an unknown sample against a handful of suspects is probably fine, but fishing in an entire database of for hits on a set of cold cases? I suspect the probability of finding a false positive grows quickly when the number of possible matches increases.

Forensic DNA use variable repeating sequences in non-protein-coding sections of DNA. Special chemicals latch onto to each key sequence. The repeat lenghth appears as distance or intensity in a chromatagraph. Enough probes and the chance of multiple matches can be less than the worlds population.

Forensic DNA was developed 30 years ago when actual DNA sequencing was mostly a pipe dream (just short disease areas).

I'm very familiar with the traditional methods of sequencing, including the commonly used forensic techniques[1]. I used to work at a well-known genetics lab way back when everybody used Sanger sequencing. I left right around when the first pyrosequencing machine was available, so I'm not familiar with the newer techniques that were developed after that point.

My question is what these newer methods are storing. Elsewhere in this threat someone mentioned that 23andme was able to inspect >100k SNPs, which would be outstanding data if they can produce it with an acceptable error rate.

> the chance of multiple matches can be less than the worlds population.

See, this is exactly what I was talking about: it depends on what question you are trying to answer. If have a sample from the scene of a crime, we can get very high confidence when matching that sample to even a large number of suspects. The "paradox" happens when you try to ask a different (but related) question that changes the search from one-to-many into many-to-many. Even with a low probability of a match, the larger number of possible matches makes false positives very common.

(if you prefer, when the prior probability is extremely low, testing entire databases will tend to find a lot of false positives)

DNA analysis is a powerful technique, but you still have to be careful to make sure you are asking the right question.

[1] there is viability in how many which unfortunately are somewhat variable, but I'm ignoring that

Do you see why Europe has data protection as a fundamental right?

I've been advocating for a US version of those data protection laws since the mid 90s. Unfortunately progress has been slow, often because of the traditional paranoia about any kind of government regulation.

As usual, we will probably have to wait until a lot of people start experiencing negative effects. Nobody cares until it hurs them personally.

The EU's data protection laws have exemptions for the prevention, detection and prosecution of criminal offenses as well as national security and defense. I don't see why law enforcement in Europe couldn't do the exact same thing as was done here.

If you look, you will find this kind of reasoning underlying a great deal of public (and increasingly, private) policy.

Women are at about 3x the risk of men for intimate partner violence, which includes sexual assault -- see http://www.cdc.gov/violenceprevention/intimatepartnerviolenc... for extensive background.

While I strongly object to mass DNA testing of the type proposed here, the DA's comment should be read in the light of the huge backlog that exists for testing rape kits. Denver, CO's clearance rate is only about 30%, and there was until recently a strong bias towards testing cases involving stranger assault but not assault from family members: http://www.endthebacklog.org/denver

So without endorsing the DA's call, it us based on a legitimate public policy issue rather than purely emotional grandstanding.

The bias against testing rape kits when a family member was the alleged rapist comes from the difficulty of prosecuting cases where both people agree that sex happened, but one says it was rape and the other says it was consensual.

There's not much point testing DNA in those cases because it doesn't help you decide if consent was gained.

(I agree with your post, btw.)

I totally agree, but one reason to do so anyway* is because of the non-zero probability that someone who attracts a rape accusation from one person may well predate upon others, sequentially or in parallel. A topical example would be Bill Cosby, who's now accused by some 50 women of sexual assault, many of whom said they kept silent for years because he was so famous and they thought nobody would believe them.

* which is fraught with its own ethical pitfalls re. DNA storage, but is in line with current US legal mores, if not your or my personal ideals on privacy.

An accusation alone is not grounds to keep your DNA information around forever. A conviction is another matter.

Playing the devils advocate here but why keep convicts' DNA forever? Should ex convicts not have a right to privacy? I can imagine they might need some way to keep track of inventory in the prison system but once you've served the time, why keep the DNA? Just to make the cops' jobs easier? If there is a false positive on an ex convict's DNA, it still means we got the wrong guy.

Since there's an actual conviction, that's a separate matter but I tend to agree.

How would DNA storage of rape kits help the Cosby case at all? The police can collect Cosby's DNA if he is a suspect in a case.

Just think about it for a second

It wouldn't. I mentioned that case in the context of (apparent) serial predation and information asymmetry.

I strongly support fixing the backlog of rape kit testing, and if I thought that's what he meant I wouldn't object. Perhaps he was quoted out of context. What I heard was a very old, very tired, very patriarchal attitude of "join me, fellow men, to protect the women and children!" I am not the kind of feminist who resents a door being held open for her, but this quote rubbed me the wrong way.

You might very well be correct - it's one big downside of DA being an elective office, which is something I'll never get used to. My mind just jumped to the rape kit problem immediately on reading your first post, so I just went with the flow of the intuition.

All but one of my 8 closest male friends have been slapped by ex-girlfriends. I don't usually see any testimonial for this behavior in statistics. There could be a strong bias against men.

Possible bias in your choice of close male friends? That looks like a little over one data point rather than eight. To counter with one data point, it has never happened to me (male).

Possible, but in the spirit of statistical rigor this could also mean that 7 of your friends dated the same violent woman.

> the DA's comment should be read in the light of the huge backlog that exists for testing rape kits

Except grabbing DNA information from private companies databases is only potentially useful IF those kits are tested. Testing the kits is orthogonal to clearing the cases.

Quite so. it would have been better for me to say 'could be read in the light of' as opposed to 'should be..'

Intimate partner violence is pretty much the least likely application I can imagine for a database of random strangers' DNA. What is the scenario? "My husband hit me in the head and I got amnesia, so I don't know who my husband is"?

FYI, that CDC survey defines rape as penetration only. They don't count forced envelopment.

I am not sure what is intimate partner violence but 40% of victims 'being hit' by their partner are men, and in most cases of 'being hit', women start it - predominantly under the assumption that no one will hit them back because a) they are women b) society still goes gaga after 'women and children' and not men c) in practice women are protected from any consequence of hitting a man but men are by default assumed guilty until proven innocent.

For example, while VAWA talks about non-gendered domestic violence, in practice only one partner is arrested so the other can be allowed to take care of children and in 100% of cases where both parties got into a physical altercation it is MEN who end up in jail and with a restraining order.

In a study from Sweden a few years ago, it showed that women and men report essentially equal amount of violence in the home, which throws a big wrench into the perception of women as the perpetual victim and men as the perpetual aggressor.

But even more interestingly the study showed that men and women choose different method in physical violence, and that has a major impact on injures. Men favors unarmed methods by the use of their hands, while women favor knives or improvised weapons. In average the unarmed method cause more often damage which need the attention of medical help, while armed combat during domestic violence causes more damage in those rare instances where medical help was needed. The explanation for this is that combat with improvised weapons is simply harder to do than unarmed.

http://www.amazon.com/Three-Languages-Politics-Arnold-Kling-...

I think the point is that many murdered women or children are often murdered by spouse or parent. The type of case which can be hard to prove. Perhaps a better quote would be "cold cases which are usually hard to prove"

Not sure why children, but in the case of women, could it be that a male perpertrator's Y chromosome is particularly easy to separate from the victim's DNA?

Gotta stop the terrorists, gotta stop the child pornographers, gotta stop the human traffickers.

Why notta?

Reply on news.ycombinator.com