[Roy78]

I've personally followed Andrew Singer's/BODE's work and have been present in his (and other forensic genealogists) presentations to the forensics community. I was literally in the audience 2 days ago for another of his. I am a professional Forensic Scientist who is specialized in DNA. My educational background is in biochemistry & molecular genetics. Feel free to ask me anything.

There is a lot of misunderstandings in this thread so far. For one, on the idea that "sperm does not automatically equal murder," is of course logically true. DNA evidence doesn't equate that point, but it does suggest that there is a putative perpetrator who could be responsible for the crime. It's not up to forensic scientists to decide who committed what crimes. We can tell that the person was "there." It's up to the detectives to argue/figure out, and ultimately the judicial system to decide the outcome/verdict.

She was raped and killed, who's DNA was on the vaginal swab? Most likely the perpetrator.

Secondly, to suggest that modern DNA forensic science is questionable is farce. There have been major issues with interpretation in the past with non-accredited crime labs/analysts/methods/persons making bogus conclusions, but these days, and for the last 15+ years, has just plain not happened. If there were unqualified interpretations they would be thrown out of the courthouse in a nanosecond.

The absoluteness of DNA mixture interpretations is getting better and better. Google "probabilistic genotyping." A human can reasonably look at data and discern multiple DNA profiles up to a few people when it comes to certain limited mixtures. Prob. Gen. software can deconvolute up to about 10 distinct people in a DNA mixture. It's basically brute force computing. I bring this up because some of the examples other's posted here happened when such tech didn't exist and bogus interpretations were going on.

As for the ethics point. I have met with many victims personally. Not one has ever not returned genuine overly emotional gratefulness of our efforts. Many brave victims become spokespeople and support for other victims. They tour my lab all the time in wonder. I can't tell you how many cases we've solved. DNA evidence, investigations, then a line up, then "that's the one!" happen all the time. These would have never happened without DNA evidence.

[nobodyandproud]

Upvoted, but even large cities continue to have problems with labs making mistakes.

That is, the science is sound but I feel private citizens (whether accused or accusing) need second and third independent verifying just to be sure.

[BugsJustFindMe]

> to suggest that modern DNA forensic science is questionable is farce.

https://strbase.nist.gov/pub_pres/Coble-ABA2014-MIX13.pdf

https://www.nytimes.com/2018/09/21/opinion/the-dangers-of-dn...

"Researchers [in 2013] from the National Institute of Standards and Technology gave the same DNA mixture to about 105 American crime laboratories and three Canadian labs and asked them to compare it with DNA from three suspects from a mock bank robbery. ... 74 labs wrongly said the sample included DNA evidence from the third suspect, an “innocent person” who should have been cleared of the hypothetical felony."

That 2013 NIST report was not published until 5 years later in 2018.

How "modern" do we need to get, exactly?

Is less than a week ago modern? https://www.nist.gov/news-events/news/2021/06/nist-publishes...

"KEY TAKEAWAY #4.3: Currently, there is not enough publicly available data to enable an external and independent assessment of the degree of reliability of DNA mixture interpretation practices, including the use of probabilistic genotyping software (PGS) systems."

> If there were unqualified interpretations they would be thrown out of the courthouse in a nanosecond.

All you have to do is search for "DNA lab mishandled" or "forensic lab mishandled" to see that what you're saying is extremely wrong and naïve. And besides this is a case where there's no courthouse, no defense, no chance to determine whether the interpretation is an error.

> There have been major issues with interpretation in the past with non-accredited crime labs/analysts/methods/persons making bogus conclusions, but these days, and for the last 15+ years, has just plain not happened.

Again see the above references to evidence mishandling and _extremely_ recent NIST reports.

You seem to suffer from the misapprehension that statistical algorithms are magic and that DNA analysis happens in a computational vacuum and that there isn't an extremely error-prone physical process involved in collecting, preparing, maintaining, and analyzing specimens and that we don't have a very long and continuing history of falsifying evidence, mishandling specimens, biased interpretation, erroneous conclusions presented as fact, and covering up procedural errors.

> She was raped and killed, who's DNA was on the vaginal swab? Most likely the perpetrator.

First, you mean whose DNA was _allegedly_ on a vaginal swab. You have no reason to believe that a specimen collected in _1956_ has been handled and preserved correctly this entire time. Second, "I think it's most likely so we don't need to worry about defense" is not how criminal justice works.

> I have met with many victims personally.

I notice that you did not say you met with the family of the accused, which is the family I was talking about. I thought that would be clear in context, but maybe not. Anyway it should be clear now.

"I have met with many victims personally" sounds a lot like you work mostly with prosecutors. Unfortunately that role often goes hand in hand with blatant disregard for exceedingly rampant and flagrant due process failures.

> We can tell that the person was "there."

You can't even reliably tell that a person's DNA is present in a mixture at the time of analysis let alone whether a person was in a particular place at a particular time 65 years ago. Believing that you can is why innocent people keep going to prison. The list of utterly bunk forensic techniques that get presented as fact because prosecution relies on juror ignorance and compliance is a mile long. Can we just stop doing that, please?