All Episodes
July 24, 2019 - Real Coffe - Scott Adams
38:01
Episode 607 Scott Adams: Othram CEO David Mittleman About Genome Crime Solving, Mueller Reruns
| Copy link to current segment

Time Text
Hey everybody!
Come on in here.
Those of you who are not glued to the Mueller testimony, which I call the Summer Reruns.
It's Mueller again.
You know what you can't get enough of?
Mueller. You know what I can't get enough of?
The Mueller Report.
We'll talk about that in a little bit, but we have a special guest who I hope will be signing in any moment so we can talk to him.
I'll introduce him in a moment.
But we have to do first things first.
And the first thing first is we have to do a little thing called the simultaneous sip.
It starts when you grab a cup or a mug or a glass.
It might be a stein, a chalice, a tanker.
It could be a thermos, maybe a flask.
Could be a canteen or even a vessel of any kind.
You could fill it with your favorite liquid.
I'm partial to coffee.
And it's because of the dopamine hit I get.
Every little sip of coffee that I enjoy with you goes straight to my brain.
Wakes me up and makes the day amazing.
Join me now for the unparalleled pleasure of the simultaneous sip.
Good stuff.
Let me do a little check in here, and it looks as though we have our guest here.
And I'm going to invite him and introduce him at the same time.
Because I can do more than one thing at a time.
David, are you here?
I'm here. Thanks for having me.
Hey, let me give you a little introduction.
We have as our guest, Dr.
David Middleman, and David is the CEO of Othrum.
Am I pronouncing that correctly?
Yes. And it's the first technology company to apply all of the power, key words all, of the power of modern sequencing and genomics in a forensics environment.
In other words, solving crimes.
And... Dr.
Middleman got his start in the Human Genome Project, so he knows what he's doing, and later received his PhD in biophysics from Baylor College of Medicine.
And you've worked in the, let's see, Baylor's Human Genome Sequencing Center, and you've been involved in some startups, and now you are the CEO of Othram, and you know your stuff.
You know your genomes.
Would that be fair to say?
Yeah, that's all I've done my whole life, just a minute.
Alright, so let's get right into the fun meat of this.
What is it that the FBI and crime organizations could do before with DNA, and what is it that Othram brings that's different and new?
Sure. So, you may have heard of CODIS. That's the set of markers that the FBI has established for identifying people.
It was built 30 years ago, and it relies on looking at 20 markers in your DNA, and if you Take those markers and mark them, then if you leave your DNA somewhere, they can make the claim that you were at that point, either at a crime scene or somewhere, and that the DNA matches because the 20 markers that you have in the crime scene match.
So, in essence, the current DNA use would be not really much more sophisticated than a fingerprint in the sense that if you don't already have somebody's fingerprint on file, it's not going to help you to catch their fingerprint at the scene because there's nothing to match it to.
Would that be a good analogy?
That's correct. The premise is that when you find somebody, you add them to the database, and that way if they commit another crime, you'll catch them again.
All right. So what do you bring that's different from that current model where they have to be in the database already or you can't catch them?
There are two things that we're bringing to the table.
The first one is, instead of looking at 20 markers, we look at the entire genome.
So there's 3 billion letters to look at, lots of markers, so we look at all of it.
The second thing that we do is we focus on using that information to either describe somebody or identify someone without having to have them in the FBI database.
So they don't need to have been previously identified to be found using this approach.
And how does that work?
How do you find somebody by their DNA if you don't already have a copy to look for?
There's a lot of things you can learn from DNA. Just to run through a brief list, first of all, just looking at your DNA sequence, you can infer where you might have been from, your ancestral lineage.
There's information That you can find in the mitochondrial sequence.
That's the part of the DNA that you get only from mom and you can trace the maternal lineage several generations back.
If you're a guy, you've also got a Y chromosome that comes only from your dad and you can follow that all the way back.
All the rest of your DNA can be used to basically match you as close as possible to parts of the world where other folks with the same DNA would have originated from.
So give me that in layman's terms.
Is that going to tell me if you're looking for a perpetrator who's Irish or Elbonian or a mixture?
It could. It could. I mean, through the history of time, country borders have changed, but certainly gross approximations of where in the world your folks or you may have come from is determinable from DNA. Okay.
And could you go so far as to tell somebody's, let's say, height?
Yes. So once we get to that point, the next thing we can do is, and just jumping back to the ancestry part, is if you know where people are from the world, You can find people that are relatives, right?
So you can find relative matches, you can find recent close and distant relatives, and then moving away from relationship testing, you can definitely determine a lot of things about someone, certainly physically.
And there's things that you'll find, for example, on tools like 23andMe, other sites that try to describe eye color, hair color, but it's gotten a lot more advanced.
My co-founder, Steve Hsu, he works in an area called Polygenic Scores.
So for a long time, we were looking for markers where One gene would cause one disease.
Sickle cell anemia is a good example.
You have one gene, one little change, and that one gene causes a very obvious change in the way you are.
You have sickle cell anemia now.
But what we've discovered is a lot of traits, including physical traits, they're determined by many genes and many markers in many genes and around genes.
And so that's why they call it polygenic.
There's lots of genetic factors that are playing a role.
And polygenic scores take sometimes tens of thousands of markers, into consideration in determining what you look like.
And so, yes, to answer your question, using polygenic scores, my co-founder, Steve Hsu, has been able to predict height, for example, down to inches.
Wow. So that's a pretty exciting area of research.
Is it possible, and I'm sure you can't quite do it today, but could you imagine a time where we could reconstruct an actual artist's rendition of a perpetrator of a crime From nothing but a good sample of DNA? So I think it would be challenging to draw someone's face from scratch from DNA,
but certainly a lot of the landmarks and a lot of the unique features on your face, many of which are used in face recognition, those probably have genetic correlates.
And so certainly you would be able to match faces to DNA sequences.
Wow. So what databases do you have available to you or does law enforcement have available to find a relative?
Because if you could find a relative Then you're narrowing it down, right?
So, hey, we found a cousin, and you can ask the cousin, hey, cousin, do you know anybody in your family who lives in Detroit?
And then you might be able to narrow it down pretty well.
But what databases does law enforcement have access to except for law enforcement's own databases?
So, a lot of folks have been using public genealogy databases.
The one that's been on the news the most is GEDmatch.
There are some other commercial and public databases that are available that have aggregated consumer data.
There's 26 million people that have been tested with consumer tests.
And that doesn't include everyone that's been tested.
There's just a consumer test like 23andMe and Ancestry.
But a lot of them will move their data to other sites so they can compare themselves.
So for example, let's say you tested at Ancestry and I tested at 23andMe and we think we might be brothers.
There's no way for us to compare because the databases are separate.
So you and I might choose to put our data into a public database and then there I can compare directly to you and say, ah, you're my brother.
You can use that same approach if you're trying to find the relationship to a sample from a person you don't know.
And it's a little more...
Oh, go ahead. Yeah, so now you have me worried.
So I have used 23andMe.
And I have checked the box that says can you see if you can connect me with any family members who might also be part of the system.
Have I, without knowing it, have I released my DNA beyond 23andMe?
You have not. So 23andMe does an exceptional job of preserving privacy and they do not allow access to their database and you are entirely safe from any public sharing if you keep your DNA within 23andMe.
But wait, hold on.
But that means that law enforcement also does not have access to it, right?
That's correct. And whether that changes in the future, who knows?
You probably remember in the 90s, internet service providers would guard their data very preciously, but after a few subpoenas, they kind of started working a little bit more.
So I don't know where the future will go, but certainly their policy is not to cooperate with law enforcement.
What about the Ancestry databases, Ancestry.com and anything like that?
Do they also protect your DNA? Ancestry.com also has a closed policy.
You can't search that database unless you're in the database.
So then what databases are there, except for databases in which I've willingly given my DNA, but they're going to protect it?
What sources do you have that law enforcement has access to?
Yeah, so again, if you and I tested a commercial database, we might choose to move our data to a public database.
One of them is called GEDmatch.
It's G-E-D-M-A-T-C-H. And that's one of a few databases that are out there that allow you to move your data willingly to a space where you can compare across companies.
Okay. At this point, are there enough people who have done that?
Let's say, what percentage of U.S. citizens would you say have at least a relative in the first or second cousin closeness who are in a database?
Would you say that we're at a point where 20% of us have at least one relative who's in an available database that law enforcement can get to, or is it closer to 80%?
Well, another way to frame the question, and hopefully I'm answering it, is like, you know, if you look for example at Northern Europeans, so people with Northern European ancestry, You probably need about 100,000 people unrelated to have a 90% chance of finding someone down to like a second or third cousin.
And so that's kind of an indirect way of answering your question.
That's an analysis that we did.
You probably need 100,000 unrelated Northern Europeans to grab 90% of US Northern Europeans to a second or third cousin.
And GEDmatch, for example, has over a million people in their database.
So I would say there's a fair chance, certainly for folks of Northern European ancestry, that you can find some connection.
Now, it's not perfect.
I mean, a third cousin is very, very distantly related, obviously.
And it doesn't guarantee that you can work a family tree and figure out who the person is.
But I think there's a reasonable chance, and I think it'll only increase.
If you look at the growth in consumer genetic testing, it's grown tremendously.
I mean, it's, you know, and there's more and more data being pushed into the public sector as well.
So the other thing that you guys do is you reconstruct bad DNA samples, right?
So if there's a crime scene or it's an old crime scene...
So how groundbreaking is your methodology for reconstructing some DNA that maybe in the old days you couldn't use at all?
Yeah, I think that's actually the exciting part for me anyways, because we know on the data side, data will increase.
It's like face recognition.
Eventually, there'll be more and more data.
The question isn't should we use it.
The question is how do we use it responsibly.
It would be no better to say we should stop doing face recognition.
That's not going to happen.
On the lab side, the DNA, as it sits, you know, DNA is pretty stable, but as it sits for decades and decades, two things happen.
Number one, it gets chemically damaged and it begins to break into little pieces.
And the smaller the pieces, the harder it is to analyze.
And the second thing is it gets mixed up with other stuff, gets contaminated.
It can be mixed with other DNA, it can be mixed with bacteria.
And so all of those things make it incredibly challenging And usually you have very little DNA, right?
It's like you have tons of blood.
There's like a little piece of a bone or, you know, a stain to be the clothing.
And so it's a real challenge to be able to get any useful information off such a small sample.
And we've done a lot on the laboratory side to do things like deplete the bacterial content so we can have more human and have the ability to pull that sequence more cost effectively.
We've done work to see if we can really, really take down to the lowest levels of limits and still, you know, very tiny quantities of DNA, but still be able to get most of the sequence back.
If we don't get all the sequence back...
Yeah, go ahead. Yeah, so we're competing with the Mueller investigation here, so I'm going to get some key...
I'll never compete with him. So let me ask some key questions, and then we'll wrap up with that.
What percentage of, let's say, violent crime scenes would you guess leave at least some DNA somewhere?
If there's violence and it's the United States and, you know, somebody got hurt, what percentage of them that, let's say, don't have direct witnesses would also have some DNA that would be useful?
Half, 20%, 80%, what would you say?
That's a good question.
I don't know. I don't know the answer to that question.
I wish I did. Okay.
Because I'm wondering, how close are we getting to a prediction I made many years ago that all major crimes would be solved between – you've got the cameras that are going to catch everything in public.
You've got DNA that will get a lot of stuff that happens behind closed doors.
Have you heard of any technology that actually sniffs?
It can actually sniff a crime scene as if it were, let's say, a mechanical nose.
Have you ever heard of that technology?
I have not. Because I was wondering if that picks up DNA. Can you pick up DNA from the air?
You can. DNA can be aerosolized, but generally these crime scenes would be, you know, the DNA would actually be on some material and be sitting there.
So you'd be basically, it'd be less sniffing and more like scraping.
So if you had, let's say you had a clean room and you introduced one naked human just as an experiment and said, walk around the room and then leave.
Do you think we would ever be able to sniff that room and determine who was there based on their DNA? We'd be able to get some information.
So, you know, the FBI talks a lot about touch DNA. So it's like you touched a doorknob or your hand touched glass.
You know, there's some oily residue left on the window or the glass.
Can you pull DNA? And the answer is probably yes.
But, you know, It may be so small that you might only get some of the DNA, right?
So getting all the DNA is going to be hard from just a little bit of a fragment.
All right. We're going to wrap up now.
Is there anything that we need to know that we did not cover?
Dr. David Middleman, CEO of Othrum.
Is there anything that we left out?
No, I think you're spot on.
I think in the next five to ten years, I think face recognition and DNA analysis will allow you to basically identify, track and resolve most violent crimes.
And so I think that's happening.
I think the most important thing we can do is number one, consent and properly explain to people what's happening when they share information about their DNA. And if there's law enforcement in this country or other countries,
I suppose, who want to contact you if they've got a cold case or some DNA and they don't know what to do with it, they would find you at Othram?
Yeah, authorum.com.
And for now, we're domestic.
We're focused on bringing value to our government.
Okay, so that's O-T-H-R-A-M. And hopefully, maybe we expose some more people to this idea.
They can find you.
Maybe we solved a few crimes today.
I appreciate it.
Thanks for having me. All right.
Thank you, David. We're going to thank you and move on.
All right. I wanted you all to hear that because...
Yeah, it's really interesting to find out where things are heading.
You know, we've got zillions of rape kits that haven't been tested, we've got all these cold crimes, we've got questions about crime everywhere, and there are some technologies that are going to take a big bite out of that if we use them right.
All right, you all want to talk about Mueller.
I turned on Mueller a little bit before I came on, and I was listening to him, and I'll tell you my first impression.
Did you have the same first impression?
That Mueller doesn't seem as competent as you hoped he would be for the type of jobs he had.
Did you all have the same reaction that...
I don't know if it's because of age.
Maybe there's age plus some specific physical problem.
Now, I don't have any reason to believe he's not mentally there.
But he's not presenting an image of...
That makes you confident, wouldn't you say?
I would say that people's confidence in his ability probably is a bit degraded.
This may also explain why he was so invisible during the entire investigation.
It seems as though he was just sort of missing in action.
Now, I just thought it was entirely because he was being professional and didn't want to talk to the press.
But maybe there was another reason.
Maybe talking to the press was a bad strategy, even if he thought it was something he should do for his job.
I would expect that there will be tons of news that comes out of this, and it will all be forgettable news.
Let me give you a preview of all the news that's going to come out of this.
Blah, blah, blah, he used these words before.
Blah, blah, blah, now he's using these other words.
We think these first words may be slightly in conflict with these other words, unless you listen to what Bill Barr said, which was interpreting his words.
But then there were his other words in the report, and then he referred to the report where his words were.
And at the end of it, you're going to say, well, I think what happened was whatever I thought was going to happen in the first place.
I thought, so people are going to come away from this with exactly the same opinion they went into it.
Most. I mean, I'm sure there'll be one or two people in the United States who changed their minds.
But I think the people who say there was nothing there are going to listen to it and say, we didn't learn anything new.
Nothing there. And the people who say there's absolutely something there are going to come away from it and say, well, I guess we showed the country there's absolutely something there.
And the only reason that Mueller did not recommend indictment is because the president was a sitting president.
And then, of course, the president will come out and say, well, no collusion, no obstruction of justice.
And by the way, when the president says no obstruction of justice, I would say that's a pretty generous interpretation of what's there.
I would rather say that what Mueller concluded was, that there's not not evidence of obstruction of justice he didn't say that there's obstruction of justice he said well there's not not evidence so that's a weird standard to say that we we can't figure out that somebody didn't do it All right,
so let's wait until the end of the Mueller thing before we can say much.
Maybe I'll pop back on here.
Maybe I won't. I don't know.
Depends what they say. Here's some more fun.
Apparently, the Justice Department is opening an investigation of the major Internet tech companies to determine whether they engage in, quote, anti-competitive practices.
How about that? So, the government has decided to maybe try to see if we could have a fair election this next time.
Because, as you know, I think it was Epstein was the expert who recently told Congress that Google alone could be moving millions of votes by manipulating the search.
There's also a new video from Project Veritas today in which A Google insider is saying that we should be worried about Google manipulating results.
The insider doesn't seem to be somebody who was personally involved with any of that stuff.
It's just somebody who is on the inside who is worried about it, same as we are.
But I don't know that they introduced any new information.
But it will look like that, so it helps the narrative.
So this is interesting.
Because here's the problem.
Between now and the election, I don't know how much the Justice Department can discover.
But I do think that the big tech companies and Google, I think, has the most to worry about.
I do think that they are probably hustling to clean up their algorithms.
So there may not be any, in the end, there may not be any kind of legal action.
But this amount of scrutiny is going to make it difficult for Google to rig the election the way they obviously want it to.
Now, what's interesting is that when you and I say, well, Google obviously wanted and continues to want to manipulate results, and they will go and do that.
But it makes you wonder if the people who work there at Google would use the same words to describe what they are obviously doing.
I think we're beyond the point of wondering if it's happening.
Is that fair to say?
Is it fair to say we've gone beyond the point of wondering if Google is manipulating results and now we're just talking about it as a fact that happened, right?
Can we all agree with that?
Now, what there is still a question about is how much difference they made and what their intentions were.
Because those are different questions.
It's fair to say that Google moved lots of votes.
It is not in evidence in a way that I would say is proven to say that they colluded to move them a certain way to get a certain outcome.
Because I think if you talk to Google, they'd say, no, all we're doing is de-emphasizing fake news.
And they could make a case on that.
I think they could totally make a case...
That they're not trying to move the election.
They're just showing less fake news.
But it would be them deciding what is fake and what is not.
So, for example, if you had a climate change denier, somebody who was an actual scientist, would their study get as much attention as the majority?
Because the majority would say climate change is a big problem.
If you had an outlier who said, well, I just did this study, It's been peer reviewed and all the good stuff, but it goes against the majority.
Would that get the same kind of play as if that study had gone the other way and confirmed or even said climate change was way worse than we thought?
Which one of those would have gotten equal play?
Well, if you're Google, you have to decide which of those two is true news and which of those is fake.
How in the world can Google possibly make that decision?
So they have two choices and neither of them are acceptable.
If I may side with Google without you hating me too much, you have to acknowledge that it is in some ways an unsolvable problem because you're asking Google to make a judgment of what's true and what's not.
Because if they just ran everything, you would get as many Holocaust deniers as you had Holocaust historians.
And then if you're searching for information and you knew nothing about this topic, you'd say, I don't know, Google has given me a lot of results to say the Holocaust didn't happen, and also a lot of results to say it did, so I don't know, I guess I can't tell if it happened or not.
Is that the world you want to live in?
No! No.
You actually need Google to make some decisions about what's true, because the alternative is even worse.
So you have two bad choices.
One bad choice is that Google doesn't make any decisions about what you see, and then you eventually just see garbage.
Garbage would be the guaranteed outcome of no filtering.
Absolute garbage, eventually.
And then if they go the other way and they do make the decisions, how do you know they're making the right ones?
How do you know? It's impossible.
How are they supposed to know what is true and what's not?
The public can't tell.
The news can't tell. Which news are they supposed to look at?
Don't you think that if Google got to decide what is real and what is not, they would just say, huh, I guess Fox News is not real news or Breitbart is not real news, so we'll just run MSNBC and CNN. Well, that would be a problem.
That would be a real problem.
How about when they talk about all these hoaxes?
How many of them do they get right, and how many of them do they get wrong?
So it's an unsolvable problem, but we should at least know what we're saying, what we're not.
Some transparency would be good.
So there's that. To my great amusement, and yours too, some clever folks have dug into the older tweets of Omar and Tlaib, the two people who were on the, quote, squad that the president referred to as maybe some people who should go back where they came from, fix that, and then come back here and tell us how to do it.
Now, the president was accused of being racially insensitive for saying, go back where you came from and then come back after you've learned how to do it, to people of color.
But it turns out that those same people of color are on record as using exactly the same language of, you know, why don't we deport you or go back to your country or whatever.
So they've used the same language in public in a political context.
Now, that should tell you that that's normal talk.
And if you are politically correct, you'd say, yes, normal talk.
It's something that if you are a person of color and you're talking to someone who isn't or you're talking to someone who's the same persuasion that you are, well, then it's okay.
But if you're President Trump and you're not a person of color and you're talking to somebody who is, well, then it's not okay.
Whose rule is that?
Well, much of society has that rule.
It's a rule that at least half of the country would say, yeah, yeah, that is a rule.
It depends who's talking.
It's only bad if the right person is saying it to the wrong person.
Then it's bad. It's not just the words are bad.
Who is it who disagrees with that interpretation of the world?
President Trump and all of his supporters?
President Trump ran on the promise, the promise, I'm going to call it a promise because I think it was as good as a promise, the promise that he would not be politically correct.
President Trump, candidate Trump, ran on an explicit campaign platform of not caring what you think about that.
And if you think about it, it might have been the smartest thing he ever did.
By simply saying, I acknowledge that political correctness is a thing.
Hear me as clearly as you can.
I'm going to act like it isn't a thing.
If you vote for me, you're going to get...
This would be President Trump, essentially.
I'm summarizing. He's basically saying, if you vote for me, you're going to get political incorrectness.
That's what I offer.
And I'm not apologizing for it.
I think it's an advantage and you should vote for it.
Then, millions of people heard his message and the totality of his message and said, yeah, yeah, I'll take that.
This package is the package I want to vote for.
And then they voted for it and then he got into office and then he did exactly what he said he was going to do.
He talked in a politically incorrect way.
What makes something politically incorrect versus just wrong?
Where's the dividing line between things that all of us would agree, okay, that's just wrong.
That's not political correctness.
That's just wrong. Where's that dividing line?
Well, one clear way you could make a dividing line is if he's saying the same sorts of things that other people are saying, and the other people are not getting in trouble for it.
That would be one way to know it.
It's a question of political correctness and not a question of the words or the ideas.
It's who's saying it and when that makes it wrong.
Political correctness. So I would say that remember, how long ago was it?
Was it 10 years ago?
Approximately 10 or 12 years ago we were talking about President Trump's tweet about some members of the squad should go back home.
Oh wait, it wasn't ten years ago.
That's right. It was one week ago?
And a week ago, it seemed like the worst thing in the world and he's done this time and this time he's gone too far.
Didn't you have all those thoughts one week ago?
And what do you think about it today?
Today you think about it as, oh, it's something they said too.
Alright, moving on. Let's talk about Mueller.
Nothing. A week ago it seemed like the entire world was on fire.
Today it seems like, oh, okay, it's just political incorrectness.
That's what he said he was going to do.
Boom. Nothing.
So, that's what we got.
Otherwise, we've got Mueller to listen to, and you all want to do that.
I can tell by my numbers on here, most of you, at least half of you have gone to watch Mueller.
I'm going to do that too.
If Mueller makes any big news, I might come back, but I'm not expecting anything that would happen.
Look like big news.
Somebody said, remember when Stormy was going to take him down?
Yeah, did you see Stormy Daniels' latest tweet about her lawyer?
Man, did that turn bad.
Talk about something turning bad for Trump's opposition.
Stormy Daniels and her They just went down in huge flaming balls of flame.
Yes, flaming balls of flame.
I just said that. So, Mueller is boring us and I think maybe making the country really wish they'd stop talking about it.
I don't think somebody says we want more DNA talk.
We want our $40 billion back.
That's right.
Okay.
Okay, nothing else going on and I'm going to let you get back to...
Oh, this has to be mentioned.
If you haven't seen this on Twitter, there was some entity that was interviewing the candidates and they talked to at least Buttigieg And at least Andrew Yang, I don't know who else they talk to, but at the end of their interview, I guess they have a thing where they say, what's your favorite curse word?
And so first they show Buttigieg being asked, what's your favorite curse word?
And apparently he didn't know the question was coming because he was a little unprepared for it.
And he just laughs and says, well, I don't think that would be a good thing to say.
So that's Buttigieg's answer is just smile and giggle and I don't think I should say that.
And then they say to Andrew Yang, what's your favorite swear word?
And Yang looks straight in the camera and he goes...
I won't say the word because you've probably got kids in the room, some of you.
But they bleep out his word.
But... I think Yang's poll ratings went up 3% just because he looked at the camera and he didn't hesitate.
He goes...
So there's no question what the word was.
It was the F word if you're listening to this and not reading my lips.
But it made me think, oh, well, there's a candidate who has at least a fighting chance of telling us the truth.
There's somebody who if you ask him a question, he might actually answer the question.
I'm trying to think of any example where anybody has yet caught Yang avoiding a question.
Have you seen that yet?
Has anybody seen Andrew Yang avoid a question?
Because I'll bet it's never happened.
But I'll bet you could find recordings of every other candidate Sort of answering a different question or saying why the question isn't fair.
So he's the most interesting candidate by far.
I still haven't made connection with somebody on his staff who was going to hook me up with an interview.
I should follow up on that.
But for now, let's go all watch Mueller and we'll talk about that later.
Export Selection