from the combine-with-automated-facial-recognition dept.
KOMO TV (Seattle) is carrying a story about unsolved "Cold Case" murders in Tacoma that occurred in 1986.
TACOMA, Wash. - Using cutting-edge technology not available until now, investigators have released composite sketches of two men suspected of abducting and killing two young Tacoma girls in 1986.
Police say they are determined to solve the two horrific murder cases, which have gone cold after three decades - and they are hopeful the new technology will help lead them to the killers.
There were no witnesses. But DNA samples were found. So how were the sketches made?
The "composite sketches" were generated by a computer based on a process called DNA Phenotyping which is the prediction of physical appearance, using information extracted from DNA which accurately predicts genetic ancestry, eye color, hair color, skin color, freckling, and face shape in individuals from any ethnic background, even individuals with mixed ancestry.
"These are composites much like a witness giving a description and a computer program making a sketch based on known appearance factors," Loretta Cool of the Tacoma police said in a prepared statement. "These composites will not be exact but the outcome is a visual reference that may look similar to what the suspects looked like in 1986."
The process was developed by Parabon Nanolabs and the process is explained on their web site.
How close are the predictions?
Parabon's website has some examples generated from DNA contributed by known volunteers. You can compare the sketches with photos of the volunteers and judge for yourself. Personally, I think Yolanda McClary's actual IMDB photo is virtually a dead ringer for the computer prediction.
Related Stories
Anonymity continues to die a little every day:
The physical traits predicted from genome sequence data may be sufficient to identify anonymous individuals in the absence of other information, according to a study set to appear in the Proceedings of the National Academy of Sciences this week.
After looking for links between physical phenotypes and whole-genome sequence data for more than 1,000 individuals from a range of ancestral groups, researchers from the US and Singapore took a crack at predicting biometric traits based on genetic data with the help of a newly developed algorithm. In a group of de-identified individuals, they reported, the algorithm made it possible to identify a significant proportion of individuals based on predictions of three-dimensional facial structure, ethnicity, height, weight, and other traits.
"By associating de-identified genomic data with phenotypic measurements of the contributor, this work challenges current conceptions of genomic privacy," senior author Craig Venter, of Human Longevity and the J. Craig Venter Institute, and his co-authors wrote. "It has significant ethical and legal implications on personal privacy, the adequacy of informed consent, the viability and value of de-identification of data, the potential for police profiling, and more."
[...] [Genome] sequences [...] are not currently protected as identifying data under the US Health Insurance Portability and Accountability Act's Safe Harbor method for ensuring anonymous and de-identified patient information.
Also at Bio-IT World, PRNewswire, and San Diego Union Tribune.
Previously: Creating Wanted Posters from DNA Samples
Related: EFF to Supreme Court: The Fourth Amendment Covers DNA Collection
Kuwait Creating Mandatory DNA Database of All Citizens, Residents--and Visitors
Massive DNA Collection Campaign in Xinjiang, China
Routine Whole Genome Sequencing: Not Scary?
The Associated Press and the Everett Washington HeraldNet carry a story about a 30 year old double murder solved using Public Genealogy Sites similar to the Golden State Killer story carried here on SoylentNews.
Deaths of two Canadian visitors shopping in the Seattle area were unsolved since 1987.
The deaths remained a mystery for more than 30 years, until DNA led to a major breakthrough. A genealogist, CeCe Moore, worked with experts at Parabon NanoLabs to build a family tree for the suspect, based on the genetic evidence recovered from the crime scenes. They used data that had been uploaded by distant cousins to public genealogy websites. They pinpointed a suspect, Talbott, a trucker living north of Sea-Tac International Airport.
Police kept him under surveillance until a paper cup fell from his truck in Seattle in early May. A swab of DNA from the cup came back as a match to the evidence that had waited 30 years. Before then, Talbott had never been considered a suspect. Days later he was in handcuffs.
This time the police used Parabon NanoLabs (more well-known for generating facial models from mere samples of DNA) to build a family tree of the killer by submitting the 30 year old crime scene DNA samples to multiple genealogy sites.
Results from those sites were combined by a Parabon genealogist to map the family of distant cousins found in those data bases. Police were then able to narrow down the list using other methods unmentioned.
Neither article mentions if any family members were stalked by police while being eliminated as suspects, or whether any samples were submitted by other family members.
Crime scene DNA could be used to reveal a suspect's age—and whether they have cancer
A drop of blood left by a suspect at a crime scene is a treasure trove for forensic scientists. Genetic information extracted from such biological samples can be compared against DNA databases to see whether a sample's DNA sequence is a match for any known offenders, for example. To protect individuals' privacy, these analyses, known as DNA fingerprinting, are normally restricted to parts of the genome not involved in creating proteins. But in some countries, investigators hoping to narrow down their pool of suspects are allowed to identify certain protein-coding sequences that can help predict skin or eye color. And soon, scientists may be able to find out even more from an offender's DNA—including their age.
A new forensic approach analyzes the chemical tags attached to DNA, rather than genetic sequences themselves. These molecules, which can switch genes on and off, get added onto DNA throughout our life span in a process called DNA methylation. And because the patterns of DNA methylation change as we age, they could provide a good indication of how old a suspect is.
But this technique could inadvertently reveal a lot more about a suspect's health and lifestyle [DOI: 10.1016/j.tig.2018.03.006] [DX], raising tricky legal and ethical questions that may demand new privacy safeguards, scientists suggest in a commentary in the July issue of Trends in Genetics.
A brief interview with two of the authors is included in TFA.
Related: Better DNA Hair Analysis for Catching Criminals
Creating Wanted Posters from DNA Samples
The Problems With DNA Evidence
Study Predicts Appearance From Genome Sequence Data
GEDmatch: "What If It Was Called Police Genealogy?"
DNA Collected from Golden State Killer Suspect's Car, Leading to Arrest
(Score: -1, Offtopic) by Anonymous Coward on Friday April 08 2016, @06:02AM
ON WEED?
(Score: -1, Offtopic) by Anonymous Coward on Friday April 08 2016, @08:19AM
indeed
(Score: 0) by Anonymous Coward on Friday April 08 2016, @07:37AM
What could possibly go wrong?
(Score: 0) by Anonymous Coward on Friday April 08 2016, @08:14AM
The criminal gets caught.
(Score: 2, Insightful) by Anonymous Coward on Friday April 08 2016, @10:12AM
The criminal gets caught.
Unless they are displaying a picture of the actual DNA - and everyone is walking around with a picture of their own DNA on it - there's a good chance that some poor schmuck is going to get dragged in by the police because he looks like a picture of someone that the police made up out of thin air. I'm sure some lawyer will be able to claim that requiring a DNA test based on a picture that isn't real, of a person that is conjured out of a computer's imagination, violates one of those pesky Amendments. And this is very different from a sketch because any witnesses will be presented with lineup cards (or an actual lineup) where they can see a collection of society's selected few that resemble the sketch. There's a big difference between taking someone's picture (or making them stand in a lineup) and forcing a DNA sample based off a picture a computer pulled out of its ass.
Can you imagine the cross examination? It will make Chief Wiggam look competent.
Defense Lawyer: Why did you bring the defendant in for questioning?
Law Enforcement: We had probable cause. He looked like the person in the wanted poster.
Defense Lawyer: And where did this picture come from? Specifically?
Law Enforcement: Um ... I think it came out of the Lieutenant's printer.
Defense Lawyer: Okay. And where did the Lieutenant's printer get it?
Law Enforcement: One of the computers made it up. I don't know how it works. The IT guys can never explain how anything works.
Defense Lawyer: Made it up? Do you often "make up" probable cause?
Law Enforcement: I don't know. The DA's office can never explain how anything works.
Defense Lawyer: Would you have arrested the Lieutenant if the unknown computer had made up a wanted poster that looked like him?
Law Enforcement: I don't think the Lieutenant would have given it to us if it looked like him.
Defense Lawyer: You don't, do you?
Law Enforcement: We usually keep those kinds of things in house.
Defense Lawyer: Obviously.
(Score: 3, Informative) by frojack on Saturday April 09 2016, @03:12AM
there's a good chance that some poor schmuck is going to get dragged in by the police because he looks like a picture of someone that the police made up out of thin air.
That happens with eye witness identifications every day, and you don't even have to even vaguely look like the actual perpetrator for that to happen.
In order to do what you say is a "good chance" the police are going to need other evidence, such as:
1) you were in the area, of the crime, and
2) your DNA would have to match the DNA used to generate the picture
Actually, you can pretty much forget number 1, if they got Number 2.
Even if you DON'T look much like the generated picture but the DNA matches you are is serious shit.
So you see, your assertion is pretty much totally bogus, because it doesn't work that way. They don't make up people out of thin air, they use DNA. Or did you not bother to read even the summary, let alone TFA?
No, you are mistaken. I've always had this sig.
(Score: 0) by Anonymous Coward on Saturday April 09 2016, @06:17AM
Creating a picture of someone base only on DNA is the same as making a picture up out of thin air. No matter how good they get at DNA "creativite imaging" it can't determine facial hair, weight, dental issues or changes, scars or tattoos, among other physical attributes. These all can and do affect how a person's face looks.
(Score: 0) by Anonymous Coward on Friday April 08 2016, @10:07AM
The comparison between photo and sketch is better than I would expect, but I wonder how well Joe/Jane Average could pick out the person sketched in a line of 10 people that look a remotely similar.
If you look at the two sketches from the people of East Asian heritage (or the Latino/European and Northern European women), you would almost say that the sketches are more or less the same, yet the photo's are completely different.
(Score: 2) by Aiwendil on Friday April 08 2016, @11:51AM
Ok, I'm bad at recognizing faces to the point where I have troubles recognizing my own mother in pictures of her face but I can often recognize if it is the same person in two images placed next to each other.
However, the example-pictures are what I would consider "not even close", this makes me wonder just how much this relies on the brains habit to fudge data in order for them to seem similar.
For instance the eyecolour, skincolour, skeletal features of cheekbones and forehead, and the dimple at the chin are all completly off. The only thing I would say this got right was the structural features of the eyes-nose-mouth and roughly if they where black/white/east-asian.
(Score: 0) by Anonymous Coward on Friday April 08 2016, @01:10PM
The right way to check it is to select a bunch of sufficiently similar looking volunteers, generate "DNA images" from them, and then give the real and computer generated images to test subjects who have to find out which generated face belongs to which real face. The accuracy of the method would then be reflected in the fraction of correct pairings, compared to the expected fraction for pure random pairings.
(Score: 2) by frojack on Saturday April 09 2016, @03:21AM
However, the example-pictures are what I would consider "not even close",
I thought they did look amazingly close, give that they were made from a cheek swab, especially if you have seen these people on TV previously. Like I mentioned, Yolanda's imdb photo is a very close match to the generated image. Kate? Not much.
They are certainly good enough to point you in the right direction when someone in the neighborhood thinks it kinda sorta looks like Joe, but totally different than Bob.
Remember they still have the DNA for an actual match.
No, you are mistaken. I've always had this sig.
(Score: 2) by Aiwendil on Saturday April 09 2016, @06:28AM
Our differences in opinion regarding if they are close or not is why I wonder just how much it relies on the brain to make a bunch of leaps.
I assume you have normal or better ability to recognize faces which [if true] would mean you identify people on different features than I do.
If you have the time I'm curious about if you still conside them similar when viewed upside down (or section-by-section when split into 1x4 or 3x3 sectiond), just to delay lots of the instinctive habits.
(Score: 1, Insightful) by Anonymous Coward on Friday April 08 2016, @12:04PM
The pictures do not look anything like the people. If you squint, wait 30 years, and the police bring you these "matching photos" then you might agree with anything they say.
I wonder how much money Parabon would get for these types of services. I should start a business that makes composite photos of "suspected terrorists". I'm sure the DHS/TSA would eat that shit up.
(Score: 0) by Anonymous Coward on Friday April 08 2016, @01:00PM
For the last example, the picture shows a blond woman, but the data under the picture says "NOT brown/red/blond (99.6% confidence)". Seems to be overly confident here.
Note that unlike the second photo, this one does not say "dyed hair", so I assume the hair colour shown is the person's natural hair colour.
(Score: 0) by Anonymous Coward on Friday April 08 2016, @01:04PM
Err … I just note my comment could be quite confusing. When I write "picture" the first time, I meant the photo, while at the second time I meant the generated image.
And just in case this causes confusion as well, the second photo is of course the photo from the second example.
(Score: 0) by Anonymous Coward on Friday April 08 2016, @04:40PM
While you would think that based on the way they labeled Kate Snow's actual photo, I'm pretty sure I see dark roots in that one.