You sound racist. And this is not the first post from you.

ur right it's definitely less reliable than a sketch artist, but I think it's still better than nothing.

Obviously shouldn't be used as evidence to convict someone, but I'm not against using it as a tool to search for someone (assuming that the police are made aware of how imperfect these renderings could be and make use of them accordingly).

for facial structure there seems to be some solid data on what genes encode which features but there are 2 major complications'

1. gene to structure models are not entirely accurate. this is problematic because facial structures are inherently similar so minor differences are actually very important to consider

2. other features such as hair, eyebrows, lips, etc. are not considered. this creates great variance in which possible features there are to look for. so the entire enterprise is essentially trashed because there is no way to account for these variables

It's not better than nothing it's useless and pretty dangerous. What ur pretty much saying is a tool that generates an image that doesn't even look like the suspect would help capture the suspect. Like what's even the point

ive seen talks on facial structure to genealogy relationships, theyre pretty interesting because the genes involved in facial structure are conserved between primates but theyre flawed in that they are mostly targeted at major features like jaw length or cheekbone location instead of more discrete measures

Hey Joe I know you dropped out of high school but here’s how to properly use this DNA a***ysis tool

Race bait gets blocked

i’m glad you did this because who???

why does it have to be entirely trashed?

why can't we say "okay, this technology will give us an idea of spacing between the eyes, forehead size, cheekbones, etc. This information will still be somewhat of a guess, but it will be more accurate than random chance." and then use this stuff as a tool?

I get that it's got plenty of shortcomings, but I don't get why that means it has to be completely trashed

that's crazy

how did they test it did a nerd in a lab coat c***on his laptop until the screen showed him a 3d rendition of his face

Because once you have narrowed down the suspects based on your best a***ysis u still got 5-10k people to consider
Or 1-3k

Either way it's too much variability to consider the model effective

put it this way

if theres a 0.001 percent chance the person did the crime based off genetic data, are u gonna prosecute them off this?

ofc not

i mean im not categorically opposed to the idea that DNA phenotyping is possible. but it's a powerful technology that needs to be presented to the public in a way that convinces them it can be, and will be, utilized as a neutral arbiter of truth.

in this situation, Edmonton PD are presenting this digital caricature, in a criminal justice context, as the truth to the public

most ppl don't have time to consider the nuances and take into account that it's a "pretty good guess." if they see a picture they're going to take it as a fact.

certainly Police Departments have a duty to shepard the normalization of such technology in a socially responsible and not disruptive way (and maybe in Edmonton they do, I don't know anything about the Edmonton police department's community outreach)

You match the description 2k22

I need you to read my posts, not just skim them, okay?

For certain facial dimensions, DNA can give us a pretty good estimate. For other facial dimensions, DNA can give us a vague, but better than random, guess. And then we have even more facial dimensions that DNA can't help us at all with.

You keep pointing out that it's not perfect, but that doesn't change the fact that there's still some useful info. Ur throwing the baby out with the bathwater, you feel?

I'd think it would be used in combination with other, stronger evidence, not all alone right?

I am reading ur posts and I'm telling you that you are wrong. No dna doesn't give a good estimate of facial dimensions or even a vague one is what I been tryna tell you for the last couple posts.

I'm not pointing out that it's just not perfect I'm pointing out that it literally does not work at this moment in time and anything that suggest otherwise is a lie

youre definitely right

in 1 out of 10 cases this would be useful evidence

but like others said, certain things are already known about the suspect (race, physical build) so its not guaranteed that this a***ysis would narrow down anything

i think its more likely to introduce inaccuracies into the process. like its more likely that the physical appearance will differ from the actual suspect than it is that the appearance will be a perfect match. ultimately its more likely to result in the wrong people being arrested than the right person being arrested because its basically false evidence that could apply to anyone

reminds me of this

so why not just use it while keeping in mind that the suspect could very well look nothing like the model?

Don't use it as evidence to convict, but do use it as a tool to identify potential suspects

we need to ban white people

Looks more accurate than saying "suspect is black between 20 and 40 between 5'1 and 6'2"

They'll probably be able to accurately recreate a person's face with just a small DNA sample eventually