4 Univ of Idaho Students Murdered, Bryan Kohberger Arrested, Moscow, Nov 2022 #90

DNA Solves
DNA Solves
DNA Solves

New Posts

Status
Not open for further replies.
Nila Aella said:
Re Two SNPs : access to the database was transferred from the private lab to the FBI.

In replying to the Judge regarding the two SNPs

Thompson says the difference is this: (8:37)

The one from the private lab is the initial profile that the private lab analyzed, created from the DNA, the portion of the DNA sample. Then what was transferred to the FBI was after that original SNP had been uploaded to a database, access to the database was transferred. And it is our understanding once a SNP is put into one of these databases, the database adjusts the SNP to fit with what their criteria are for them to do their searches. So there are differences between the entirety of the two SNPs.

https://twitter.com/x/status/1720179509746622837

ETA: JMO
Edit to change Othram to private lab
Click to expand...
Thanks for transcribing that part for us!!!
 
10ofRods said:
Do you understand what an SNP profile is and why there would be multiple profiles? (It's a profile, an algorithm). I have tried to explain it several times on these threads, but will do so again if someone asks. If you know what it means, then it's no surprise that the two profiles are different - they have to be. I assume you're just stating the facts rather than asking a question.

...snipped by me
Click to expand...
I think you are missing the point. I'm not questioning the SNP profile. I'm curious about exactly WHAT is "altered" and HOW it is altered by the database software. In other words, HOW does the SNP data from the lab look BEFORE going into the database and once the database alters it to fit the requirements of their software what does it look like then? I'm not talking about the output, I'm talking about the machine language. What do you know about that?
 
Balthazar said:
I think you are missing the point. I'm not questioning the SNP profile. I'm curious about exactly WHAT is "altered" and HOW it is altered by the database software. In other words, HOW does the SNP data from the lab look BEFORE going into the database and once the database alters it to fit the requirements of their software what does it look like then? I'm not talking about the output, I'm talking about the machine language. What do you know about that?
Click to expand...

But, let's say for argument sake that it alters it completely. Devil's advocate, they built the tree on this "altered" profile, received an investigative lead, still did not arrest him but instead surveilled him, and witnessed what they did (him with the gloves at the store, cleaning the car, throwing his trash in the neighbor's bins). They took a sample from the family trash. It came back to statistically being the biological father of the perpetrator who left his DNA on the sheath at the scene. With that evidence, combined at that point with video evidence, cell phone data, DMV records, and more as evidenced in the PCA, they swore out the arrest warrant and retrieved his direct DNA either on-site in his kitchen or at the precinct (the warrant return of his person doesn't tell us where they were but I think that's irrelevant. We know it was on arrest). That direct DNA comes back as a match to the cs DNA.

How does your altered SNP profile help him? I think you are putting too much stock into "altered". I don't think they intend it the way you are receiving it.

jmo
 
A "Final End?
10ofRods said:
.... I doubt that there has ever been a "final end" to any legal issue...since Roman times and good records have been kept. ...
Click to expand...
snipped for focus @10ofRods

Speaking generally to legal issues and not just to this PC or BARD issue
^^^^^^^^^ very wise words.^^^^^^^^
 
Last edited:
Jurisprudence said:
But, let's say for argument sake that it alters it completely. Devil's advocate, they built the tree on this "altered" profile, received an investigative lead, still did not arrest him but instead surveilled him, and witnessed what they did (him with the gloves at the store, cleaning the car, throwing his trash in the neighbor's bins). They took a sample from the family trash. It came back to statistically being the biological father of the perpetrator who left his DNA on the sheath at the scene. With that evidence, combined at that point with video evidence, cell phone data, DMV records, and more as evidenced in the PCA, they swore out the arrest warrant and retrieved his direct DNA either on-site in his kitchen or at the precinct (the warrant return of his person doesn't tell us where they were but I think that's irrelevant. We know it was on arrest). That direct DNA comes back as a match to the cs DNA.

How does your altered SNP profile help him? I think you are putting too much stock into "altered". I don't think they intend it the way you are receiving it.

jmo
Click to expand...

Well stated. It appears both were offered to be completely transparent as well as thorough. This is some nuanced information and very challenging to understand for lay people. JMOO

I read it as here's the original and here is the SNP from our process, thereby making it different and you could use the word altered but it comes out the other end in a different format because of their process not because they intended to alter it.
 
Balthazar said:
I think you are missing the point. I'm not questioning the SNP profile. I'm curious about exactly WHAT is "altered" and HOW it is altered by the database software. In other words, HOW does the SNP data from the lab look BEFORE going into the database and once the database alters it to fit the requirements of their software what does it look like then? I'm not talking about the output, I'm talking about the machine language. What do you know about that?
Click to expand...

I have answered it every way I know how. The original data is not altered. It's just there. The original files and the original DNA is still there (in this case, on swabs that are preserved).

You are missing the point. Nothing is altered. Whose word is that, even? Did a geneticists say it? No, they did not.

If I draw a picture of you, I do not alter you. You are still there. Same with PROFILES. They are snapshots of the genome and each photographer does it differently. All are based on reality. Reality is not altered. DNA is not altered by the profiling process. In this case, it's also safely preserved in BK's body, as well as in the lab replication/analysis. It has been COPIED but not altered (altering DNA is an incredibly difficult thing to do requiring advanced technologies beyond the reach of any forensic lab).

But labs like Parabon and Othram never see that actual real DNA. They don't have blood from people. They don't even have saliva (well, they can collect it - but 90 percent of what they do is based on people's results from Ancestry and 23andme - where you do pay money to have your saliva analyzed).

The database is not altered. What database do you even mean? If you get blood test results that show you are anemic, is that result a "database"? DNA is a long string of graphical points (or letters, used as abbreviations). It's like a giant book that can be read by knowledgeable geneticists who then use computer programs to aid their reading (because no one understands the whole genome - each of these groups we're talking about is only studying a tiny fraction of the data).

All they are doing is basically putting different lenses on a kind of mental microscope. When you look through a microscope, nothing is altered in the slide (if you follow scientific procedures).

Nothing in the data is altered. The DNA submitted to Othram for the match (Kohberger's dad's dna) is still precisely what it was before (both physically and in its representation).

I can put it a slightly different way. In order to analyze SNP's, we MUST use computers. Some locations in the genome have up to 6000 different alleles that could be in the slot (and more are being found continually). But all Othram or Gedmatch are trying to do is MATCH one person's DNA (without any regard to what is being matched) to another person's DNA. In order to make this fast and not take years to get results, they FOCUS their lens on 600,000 SNP's (that's a lot). These are individual base pairs in the genome.

The FBI lens FOCUSES on only 18 (producing way more matches - as intended, as their goal is to continually scrutinize already known felons to make sure they're not committing new crimes). They don't care if they have to go to some guy on probation and make him account for where he was. They like doing that. I may be wrong about the exact number (it used to be 16, then it was 18, I haven't checked this year).

You want a MACRO lens for the FBI person, and you want a MICRO lens for matching two different people. You're playing a card game. One game has 18 cards, the other has 600,000 cards. The total deck is much bigger. The deck never changes (except by mutations in sex cells - which happen every generations but are of no value whatsoever in making familial connections, because that's the one region of the individual's genome where they will likely match no one).

If you take a snapshot of what you seem to want to call a "database" (to me it's not a database if it's just data - it's a database when you have individual files - and those individual files in the database CANNOT be altered or rewritten - that's fraudulent, unscientific and further, it would only change the letters in one version of a file).

The REAL Kohberger genome is now in the possession of the State of Idaho in several forms. Someone looking at the letters/graphical points that represent that genome, on a computer, at the other side of the country, is NOT altering any data. Couldn't get their work done properly if they did. They are simply looking for matches, just like that kids' game. And since it's super tedious and boring work, they use computers to do it (faster) for them.

My computer is not altering its basic software as I type. Nothing I type in this text box will alter my operating system.

Anyway - what do you even mean by "database." Can you say what words you use to refer to actual DNA? And to an actual print-out of a human's DNA? I would say "DNA" and "DNA results." The ISL said they were 'lucky' to find enough complete DNA on the use point of the sheath (a common place to find complete DNA) that they could run several swabs. Each gave a complete human genome - they just didn't know who it was.

So are you claiming the lab altered the RECORD of the DNA (for what reason? why do you think that? I understand general mistrust of science, if that's your reason, but of course, I can't really have hope for a sensible conversation about it, if that's your plan).

OR, are you thinking that they altered the DNA itself? (Which I am telling you is not possible for the lab to do). Genetic engineering is a completely different thing. And since we now know that the DNA actually DID match Bryan Kohberger, we also know that from start to finish, the DNA was complete; it was compared to many other people's DNA at Othram (match game) and got matches. One of the closer matches (all partial of course, as BK wasn't in that database as a known individual) was a Kohberger, which immediately rang a bell to the investigators, as he was already on their radar.

Then they got Dad's DNA, ran it against the BK DNA (already run by ISL; with its record being entirely digital - NO ONE can actually see DNA with their eyes - except using electronic microscopes or similar, but even then, only biochemical analysis shows the SNP's). Dad's DNA matched at about 50% (proof of paternity - no one else in your world or mine is going to match you at 50% except your parents; keep in mind that these types of matches are ultimate run using the entire long set of graph points/letters (nucleotides) from the original DNA.

Database software. It compares. It does not alter. It wouldn't work (at all) if it altered and Othram and Gedmatch and 23andme and Ancestry would all go out of business. It just proves a READER that a human can use to understand - unlike the REAL results, which take years and years and teams of scientists to begin to ponder.

Here is one graphical representation of those base pairs (SNPs), and this is how it looks when we do it in the lab (this is a common staining technique under electron micrscope and represents a FRACTION of a person's actual DNA):

1699132011994.png

That's a stock photo from Getty. As you can see, you can't make heads or tails of it. Nor can I. This is just one computer-assisted way of lining up the original DNA results. But to actually multiple this picture by 1000 and then expect a human to figure out comparisons is impossible. We can't do it.

And here's a picture of the results of a paternity test:

1699132096962.png


Now, I know that those colors represent nucleotides, introns and extrons. So does the computer. In this case, we have selected a subset of the original data to compare (SNP's which are the only places that two individuals vary). Each of those colors represents a long list of even smaller data points (ACTG and more).

When we ask the computer to compare the above picture with someone else's paternity test, it will take us a method and quite some time to be able to see the differences (and similarities). It will not leap out at us. There's too many data points. Although, I've watched people work (and you can see Spencer Wells do it in his film Journey of Man, free on Youtube, at about the 45 minute mark - when he's in India). He's an expert in one chromosome. He's able to look at something like this (he prefers mathematical representations) and quickly see certain variations he's looking for. Page 216 of the below article shows how one single element in the chart above (purine) can be represented mathematically in relationship to other parts of DNA:


Keep reading to get to how to look at ONE gene mathematically (it's pages and pages - hopefully you can get a sense of what the computer is actually doing - it's doing math, quickly).

Here's what ONE gene (the gene that makes our blood red) looks like across species in its basic form:

Table 2. Listing of the bases of the first exon in the beta globin gene for the eight speciesmentioned. (Note: All the papers have used 90 bases for the rabbit exon 1 but it should be 92bases. Here we report the corrected sequence.)HUMAN (92 bases):ATGGTGCACCTGACTCCTGAGGAGAAGTCTGCCGTTACTGCCCTGTGGGGCAAGGTGAACGTGGATGAAGTTGGTGGTGAGGCCCTGGGCAG
GOAT (86 bases):ATGCTGACTGCTGAGGAGAAGGCTGCCGTCACCGGCTTCTGGGGCAAGGTGAAAGTGGATGAAGTTGGTGCTGAGGCCCTGGGCAG
OPOSSUM (92 bases):ATGGTGCACTTGACTTCTGAGGAGAAGAACTGCATCACTACCATCTGGTCTAAGGTGCAGGTTGACCAGACTGGTGGTGAGGCCCTTGCCAG
GALLUS (92 bases):ATGGTGCACTGGACTGCTGAGGAGAGGCAGCTCATCACCGGCCTCTGGGGCAAGGTCAATGTGGCCGAATGTGGGGCCGAAGCCCTGGCCAG
LEMUR (92 bases):ATGACTTTGCTGAGTGCTGAGGAGAATGCTCATGTCACCTCTCTGTGGGGCAAGGTGGATGTAGAGAAAGTTGGTGGCGAGGCCTTGGGCAG
MOUSE (92 bases):ATGGTGCACCTGACTGATGCTGAGAAGGCTGCTGTCTCTTGCCTGTGGGGAAAGGTGAACTCCGATGAAGTTGGTGGTGAGGCCCTGGGCAG
RABBIT (92 bases):ATGGTGCATCTGTCCAGTGAGGAGAAGTCTGCGGTCACTGCCCTGTGGGGCAAGGTGATTGTGGAAGAAGTTGGTGGTGAGGCCCTGGGCAG
RAT (92 bases):ATGGTGCACCTAACTGATGCTGAGAAGGCTACTGTTAGTGGCCTGTGGGGAAAGGTGAACCCTGATAATGTTGGCGCTGAGGCCCTGGGCAG

How quick are you at finding similarities there? I find the letters far easier to compare than the colors, but that's just me. Are you noticing all the similarities?

A computer that can mathematically deduce not only the comparisons but the actual biochemistry that these letters are indicating is the main tool for analysis. I noticed immediately that the lemur gene was closer to human than the mouse.

Analysis with a computer does not change the original data. It too knows that the first three letters of this particular gene (and the last three - they're call CODONS, and they are read by the body to produce every single thing that keeps you alive and makes your body what it is) are the same. Hmmm. So, across several species, there are commonalities in this gene. But what about the biochemical similarity? THere are multiple codons for each amino acid (and some don't code for amino acids at all). The computer knows this system and can immediately give the formula for the protein (and it better make hemoglobin!! And - it turns out, all of these genes DO make hemoglobin, through very similar biochemical processes).

50 years ago, we'd have only said, "it's hemoglobin, from a mammal." Now we can tell if it's Lemur hemoglobin or Human (and I have no clue what a Gallus is, maybe a type of rabbit?) Anyway, we now know that our closest living relatives of the group studied here is...well, can you figure it out? Or would you prefer that someone else ran this through an analyzer and computed the actual similarities and differences?

So that's what we're up to, in our labs. My work has to do with how to develop field methods for studying these things out in the real world. In humans. I don't study rats or lemurs. And I know quite a bit about the human genome, after 50 years in the field - BUT, I know almost nothing about any one particular SNP, because each of them would require me to take about a decade of the equivalent of post-doctoral study to be familiar enough to have a conversation with the people who are studying this.

Othram hires experts in genomic study PLUS computer experts and has managed to build one of the more robust SNP study databases (the datacomes from papers like the ones I'm quoting - that's the database). The incoming DNA from subjects is then compared to it.

(Sorry for the very long post - everyone here is so intelligent, but, we were born at different times - I know there is one younger WSer who already knows more about all this than I do, but may not have years of trying explain these methods both to grant writers, to students, to LE and to the general public). She may not have the patience either - I really feel it's important in True Crime for all of us to *try* and understand the science.
 
P.S. I hate to criticize Thompson and his method of explaining (he's getting there) but yes, he's struggling but he didn't want to use all the pixels I just used.

SNP's are derived from a larger dataset. Very few projects call for studying just one SNP.

FBI = MACRO (largest possible net over a large criminal population)

OTHRAM = MICRO (huge number of SNP's under study - made possible only recently through the vast expansion of genetic research). See the pdf in my post above - if you really want to get down into the technical representation of DNA.
 
Thank you 10ofRods. I understood Balthazar’s question as well. I had a math teacher once that maintained, “it doesn’t matter how you the answer, as long as it’s correct.” Yet I would remain curious of the process.
It’s good to know, and very well explained now: there is no “altering”. I think the first time I saw that word, I thought of carbon dating where the actual sample is destroyed.
I’d like to thank everyone for their comments, questions and answers which has taught me much of law and investigating. Never been on here before November last year, but the loss, finality and brutality of this case really piqued my interest. This site and all you commenters bring that feeling of a crucial tournament; most of us are on the same side rooting for legal justice for the kids.
 
I think the confusion comes in for the same reason that people who submit their saliva to both Ancestry and 23 are upset when they get "different results." But in fact, the results are quite similar and only vary in percentages.

SNP's are not all agreed upon. How many need to match to show ancestry? Ancestry has one view, 23 has another. I've had my results changed on 23andme 3 times over the years (only submitted once) but they're using different algorithms. And that's often due to new findings in science. For example, a rare hemoglobin SNP found among the Yakuts people (Siberia) was once thought to be entirely unique to them, and so they showed up on people's 23andme results if one had that same marker.

Since then, it's been found that this rare marker occurred across Northern Eurasia among people living 28,000 years ago. So I went from being "1% Yakuts" to being "1 % Ancient Indo-European." And of course, these percentages mean very little at that distance - it just means I have that allele. Yakuts people have probably had it longer, but most people with it are from Indo-European speaking groups at 28,000 years ago. (So did some small tribe of Yakuts people valiantly become some of the early populators of Europe??)

Ancestry never even used that marker - ever. It sticks very closely to the markers that are its main bread and butter - the markers that show up in the all the genealogically driven submitters who are closely related to each other.

23 uses more SNP's per ethnic group/ancestry group than Ancestry. Both are able to give clues - but really, you need to sit down with your own family tree and realize that all your genes (including that odd one that shows up in genetic testing) came from your own ancestors, and not anyone else's. With some historical research and common sense, one can understand why they have a dash of this or that (or, by contrast, come back as nearly 100% of one ancestral group). Europe is divided into all these small nations and subregions, most of whom have ancestral records to use to plot all this out. Highland New Guinea does not. Ancestry isn't very good at New Guinea markers - but the woman who founded 23andme was very interested in the work of Luca Cavalli-Sforza, and that's exactly what he likes to do (find isolated tribes and test for markers - and 23 has been pretty good about incorporating all that side of the research).

However, 23andme has a commercial purpose and they can't keep revising people's ancestry, it makes some people really irritated.

I assume it was the Defense who used the word "altered." What gets altered is our perspective, just like using a microscope vs a telescope. Not the actual nature of the physical evidence or universe.
 
10ofRods said:
I have answered it every way I know how. The original data is not altered. It's just there. The original files and the original DNA is still there (in this case, on swabs that are preserved).

You are missing the point. Nothing is altered. Whose word is that, even? Did a geneticists say it? No, they did not.

If I draw a picture of you, I do not alter you. You are still there. Same with PROFILES. They are snapshots of the genome and each photographer does it differently. All are based on reality. Reality is not altered. DNA is not altered by the profiling process. In this case, it's also safely preserved in BK's body, as well as in the lab replication/analysis. It has been COPIED but not altered (altering DNA is an incredibly difficult thing to do requiring advanced technologies beyond the reach of any forensic lab).

But labs like Parabon and Othram never see that actual real DNA. They don't have blood from people. They don't even have saliva (well, they can collect it - but 90 percent of what they do is based on people's results from Ancestry and 23andme - where you do pay money to have your saliva analyzed).

The database is not altered. What database do you even mean? If you get blood test results that show you are anemic, is that result a "database"? DNA is a long string of graphical points (or letters, used as abbreviations). It's like a giant book that can be read by knowledgeable geneticists who then use computer programs to aid their reading (because no one understands the whole genome - each of these groups we're talking about is only studying a tiny fraction of the data).

All they are doing is basically putting different lenses on a kind of mental microscope. When you look through a microscope, nothing is altered in the slide (if you follow scientific procedures).

Nothing in the data is altered. The DNA submitted to Othram for the match (Kohberger's dad's dna) is still precisely what it was before (both physically and in its representation).

I can put it a slightly different way. In order to analyze SNP's, we MUST use computers. Some locations in the genome have up to 6000 different alleles that could be in the slot (and more are being found continually). But all Othram or Gedmatch are trying to do is MATCH one person's DNA (without any regard to what is being matched) to another person's DNA. In order to make this fast and not take years to get results, they FOCUS their lens on 600,000 SNP's (that's a lot). These are individual base pairs in the genome.

The FBI lens FOCUSES on only 18 (producing way more matches - as intended, as their goal is to continually scrutinize already known felons to make sure they're not committing new crimes). They don't care if they have to go to some guy on probation and make him account for where he was. They like doing that. I may be wrong about the exact number (it used to be 16, then it was 18, I haven't checked this year).

You want a MACRO lens for the FBI person, and you want a MICRO lens for matching two different people. You're playing a card game. One game has 18 cards, the other has 600,000 cards. The total deck is much bigger. The deck never changes (except by mutations in sex cells - which happen every generations but are of no value whatsoever in making familial connections, because that's the one region of the individual's genome where they will likely match no one).

If you take a snapshot of what you seem to want to call a "database" (to me it's not a database if it's just data - it's a database when you have individual files - and those individual files in the database CANNOT be altered or rewritten - that's fraudulent, unscientific and further, it would only change the letters in one version of a file).

The REAL Kohberger genome is now in the possession of the State of Idaho in several forms. Someone looking at the letters/graphical points that represent that genome, on a computer, at the other side of the country, is NOT altering any data. Couldn't get their work done properly if they did. They are simply looking for matches, just like that kids' game. And since it's super tedious and boring work, they use computers to do it (faster) for them.

My computer is not altering its basic software as I type. Nothing I type in this text box will alter my operating system.

Anyway - what do you even mean by "database." Can you say what words you use to refer to actual DNA? And to an actual print-out of a human's DNA? I would say "DNA" and "DNA results." The ISL said they were 'lucky' to find enough complete DNA on the use point of the sheath (a common place to find complete DNA) that they could run several swabs. Each gave a complete human genome - they just didn't know who it was.

So are you claiming the lab altered the RECORD of the DNA (for what reason? why do you think that? I understand general mistrust of science, if that's your reason, but of course, I can't really have hope for a sensible conversation about it, if that's your plan).

OR, are you thinking that they altered the DNA itself? (Which I am telling you is not possible for the lab to do). Genetic engineering is a completely different thing. And since we now know that the DNA actually DID match Bryan Kohberger, we also know that from start to finish, the DNA was complete; it was compared to many other people's DNA at Othram (match game) and got matches. One of the closer matches (all partial of course, as BK wasn't in that database as a known individual) was a Kohberger, which immediately rang a bell to the investigators, as he was already on their radar.

Then they got Dad's DNA, ran it against the BK DNA (already run by ISL; with its record being entirely digital - NO ONE can actually see DNA with their eyes - except using electronic microscopes or similar, but even then, only biochemical analysis shows the SNP's). Dad's DNA matched at about 50% (proof of paternity - no one else in your world or mine is going to match you at 50% except your parents; keep in mind that these types of matches are ultimate run using the entire long set of graph points/letters (nucleotides) from the original DNA.

Database software. It compares. It does not alter. It wouldn't work (at all) if it altered and Othram and Gedmatch and 23andme and Ancestry would all go out of business. It just proves a READER that a human can use to understand - unlike the REAL results, which take years and years and teams of scientists to begin to ponder.

Here is one graphical representation of those base pairs (SNPs), and this is how it looks when we do it in the lab (this is a common staining technique under electron micrscope and represents a FRACTION of a person's actual DNA):

View attachment 458239

That's a stock photo from Getty. As you can see, you can't make heads or tails of it. Nor can I. This is just one computer-assisted way of lining up the original DNA results. But to actually multiple this picture by 1000 and then expect a human to figure out comparisons is impossible. We can't do it.

And here's a picture of the results of a paternity test:

View attachment 458240


Now, I know that those colors represent nucleotides, introns and extrons. So does the computer. In this case, we have selected a subset of the original data to compare (SNP's which are the only places that two individuals vary). Each of those colors represents a long list of even smaller data points (ACTG and more).

When we ask the computer to compare the above picture with someone else's paternity test, it will take us a method and quite some time to be able to see the differences (and similarities). It will not leap out at us. There's too many data points. Although, I've watched people work (and you can see Spencer Wells do it in his film Journey of Man, free on Youtube, at about the 45 minute mark - when he's in India). He's an expert in one chromosome. He's able to look at something like this (he prefers mathematical representations) and quickly see certain variations he's looking for. Page 216 of the below article shows how one single element in the chart above (purine) can be represented mathematically in relationship to other parts of DNA:


Keep reading to get to how to look at ONE gene mathematically (it's pages and pages - hopefully you can get a sense of what the computer is actually doing - it's doing math, quickly).

Here's what ONE gene (the gene that makes our blood red) looks like across species in its basic form:

Table 2. Listing of the bases of the first exon in the beta globin gene for the eight speciesmentioned. (Note: All the papers have used 90 bases for the rabbit exon 1 but it should be 92bases. Here we report the corrected sequence.)HUMAN (92 bases):ATGGTGCACCTGACTCCTGAGGAGAAGTCTGCCGTTACTGCCCTGTGGGGCAAGGTGAACGTGGATGAAGTTGGTGGTGAGGCCCTGGGCAG
GOAT (86 bases):ATGCTGACTGCTGAGGAGAAGGCTGCCGTCACCGGCTTCTGGGGCAAGGTGAAAGTGGATGAAGTTGGTGCTGAGGCCCTGGGCAG
OPOSSUM (92 bases):ATGGTGCACTTGACTTCTGAGGAGAAGAACTGCATCACTACCATCTGGTCTAAGGTGCAGGTTGACCAGACTGGTGGTGAGGCCCTTGCCAG
GALLUS (92 bases):ATGGTGCACTGGACTGCTGAGGAGAGGCAGCTCATCACCGGCCTCTGGGGCAAGGTCAATGTGGCCGAATGTGGGGCCGAAGCCCTGGCCAG
LEMUR (92 bases):ATGACTTTGCTGAGTGCTGAGGAGAATGCTCATGTCACCTCTCTGTGGGGCAAGGTGGATGTAGAGAAAGTTGGTGGCGAGGCCTTGGGCAG
MOUSE (92 bases):ATGGTGCACCTGACTGATGCTGAGAAGGCTGCTGTCTCTTGCCTGTGGGGAAAGGTGAACTCCGATGAAGTTGGTGGTGAGGCCCTGGGCAG
RABBIT (92 bases):ATGGTGCATCTGTCCAGTGAGGAGAAGTCTGCGGTCACTGCCCTGTGGGGCAAGGTGATTGTGGAAGAAGTTGGTGGTGAGGCCCTGGGCAG
RAT (92 bases):ATGGTGCACCTAACTGATGCTGAGAAGGCTACTGTTAGTGGCCTGTGGGGAAAGGTGAACCCTGATAATGTTGGCGCTGAGGCCCTGGGCAG

How quick are you at finding similarities there? I find the letters far easier to compare than the colors, but that's just me. Are you noticing all the similarities?

A computer that can mathematically deduce not only the comparisons but the actual biochemistry that these letters are indicating is the main tool for analysis. I noticed immediately that the lemur gene was closer to human than the mouse.

Analysis with a computer does not change the original data. It too knows that the first three letters of this particular gene (and the last three - they're call CODONS, and they are read by the body to produce every single thing that keeps you alive and makes your body what it is) are the same. Hmmm. So, across several species, there are commonalities in this gene. But what about the biochemical similarity? THere are multiple codons for each amino acid (and some don't code for amino acids at all). The computer knows this system and can immediately give the formula for the protein (and it better make hemoglobin!! And - it turns out, all of these genes DO make hemoglobin, through very similar biochemical processes).

50 years ago, we'd have only said, "it's hemoglobin, from a mammal." Now we can tell if it's Lemur hemoglobin or Human (and I have no clue what a Gallus is, maybe a type of rabbit?) Anyway, we now know that our closest living relatives of the group studied here is...well, can you figure it out? Or would you prefer that someone else ran this through an analyzer and computed the actual similarities and differences?

So that's what we're up to, in our labs. My work has to do with how to develop field methods for studying these things out in the real world. In humans. I don't study rats or lemurs. And I know quite a bit about the human genome, after 50 years in the field - BUT, I know almost nothing about any one particular SNP, because each of them would require me to take about a decade of the equivalent of post-doctoral study to be familiar enough to have a conversation with the people who are studying this.

Othram hires experts in genomic study PLUS computer experts and has managed to build one of the more robust SNP study databases (the datacomes from papers like the ones I'm quoting - that's the database). The incoming DNA from subjects is then compared to it.

(Sorry for the very long post - everyone here is so intelligent, but, we were born at different times - I know there is one younger WSer who already knows more about all this than I do, but may not have years of trying explain these methods both to grant writers, to students, to LE and to the general public). She may not have the patience either - I really feel it's important in True Crime for all of us to *try* and understand the science.
Click to expand...

Thank you very much for your amazing discussions of genetics.

By the way, ‘gallus’ in this context should be the domestic chicken. ;)
 
10ofRods said:
IANAL, but I think it's because the actual proximate cause of the warrant had nothing to do with the IGG database. It was used to find "Kohbergers."

They used the DNA to find "A Kohberger" (not Bryan Kohberger).

Then, LE investigators saw the triangulation ( deduction, not a piece of DNA). They then went and got DNA from MK (BK's dad).

THAT is likely part of the process. MK only had 1 son.

And so, the only possible suspect was BK. Logic plus science.

I guess that might be a bit new to the legal system, but it has adjusted before. As a society, I believe we will be better off if we use our new knowledge of genes to our best advantage. If we decide, as humans, not to progress in our use and knowledge gained from science, it's on us.

And that's what AT and her client are apparently trying to exploit. Leave both logic and science out of this case, and there is no case. That's what the arguments (and costs to Idaho) are about.

We shall see what happens. If the Courts throw out this kind of investigation, then my victim-friendly heart is pained to a degree I cannot express.

But I do not think that's what's going to happen. I think we now have many, many people who understand DNA and its relevance.

I hope.
Click to expand...
MOO - For some of us, it’s all thanks to you, whom explains it so succinctly for us laypeople that otherwise would still be completely lost & confused. I, for one, thank you!
 
10ofRods said:
Great post. I too think he's enjoying his role as a jailhouse lawyer - which in some ways, he already was. He was intensely interested in crime.

I think he's been in a kind of solitary confinement all his life, unable to connect to others, seeing very little purpose to life - except, somehow, a life of crime. If he's helping others, it's possible he is getting limited time with others during his one hour a day out of his cell. It's possible he now gets more time than that - I doubt they'll keep us updated.

I know that they can use a tablet for limited communication - I wonder if they are able to communicate with each other using devices. Wouldn't be surprised. But if he's helping others, he's getting *some* contact with others - and with the power dynamic (he's the knowledgeable one, the others need help) that he prefers. TBF, it's a dynamic inherent in most jobs - so he's given himself a purpose.

IMO.
Click to expand...
Yes, the popular guy on 'campus now, no longer the outcast. I should think he'd love it.
 
10ofRods said:
I have answered it every way I know how. The original data is not altered. It's just there. The original files and the original DNA is still there (in this case, on swabs that are preserved).

You are missing the point. Nothing is altered. Whose word is that, even? Did a geneticists say it? No, they did not.

If I draw a picture of you, I do not alter you. You are still there. Same with PROFILES. They are snapshots of the genome and each photographer does it differently. All are based on reality. Reality is not altered. DNA is not altered by the profiling process. In this case, it's also safely preserved in BK's body, as well as in the lab replication/analysis. It has been COPIED but not altered (altering DNA is an incredibly difficult thing to do requiring advanced technologies beyond the reach of any forensic lab).

But labs like Parabon and Othram never see that actual real DNA. They don't have blood from people. They don't even have saliva (well, they can collect it - but 90 percent of what they do is based on people's results from Ancestry and 23andme - where you do pay money to have your saliva analyzed).

The database is not altered. What database do you even mean? If you get blood test results that show you are anemic, is that result a "database"? DNA is a long string of graphical points (or letters, used as abbreviations). It's like a giant book that can be read by knowledgeable geneticists who then use computer programs to aid their reading (because no one understands the whole genome - each of these groups we're talking about is only studying a tiny fraction of the data).

All they are doing is basically putting different lenses on a kind of mental microscope. When you look through a microscope, nothing is altered in the slide (if you follow scientific procedures).

Nothing in the data is altered. The DNA submitted to Othram for the match (Kohberger's dad's dna) is still precisely what it was before (both physically and in its representation).

I can put it a slightly different way. In order to analyze SNP's, we MUST use computers. Some locations in the genome have up to 6000 different alleles that could be in the slot (and more are being found continually). But all Othram or Gedmatch are trying to do is MATCH one person's DNA (without any regard to what is being matched) to another person's DNA. In order to make this fast and not take years to get results, they FOCUS their lens on 600,000 SNP's (that's a lot). These are individual base pairs in the genome.

The FBI lens FOCUSES on only 18 (producing way more matches - as intended, as their goal is to continually scrutinize already known felons to make sure they're not committing new crimes). They don't care if they have to go to some guy on probation and make him account for where he was. They like doing that. I may be wrong about the exact number (it used to be 16, then it was 18, I haven't checked this year).

You want a MACRO lens for the FBI person, and you want a MICRO lens for matching two different people. You're playing a card game. One game has 18 cards, the other has 600,000 cards. The total deck is much bigger. The deck never changes (except by mutations in sex cells - which happen every generations but are of no value whatsoever in making familial connections, because that's the one region of the individual's genome where they will likely match no one).

If you take a snapshot of what you seem to want to call a "database" (to me it's not a database if it's just data - it's a database when you have individual files - and those individual files in the database CANNOT be altered or rewritten - that's fraudulent, unscientific and further, it would only change the letters in one version of a file).

The REAL Kohberger genome is now in the possession of the State of Idaho in several forms. Someone looking at the letters/graphical points that represent that genome, on a computer, at the other side of the country, is NOT altering any data. Couldn't get their work done properly if they did. They are simply looking for matches, just like that kids' game. And since it's super tedious and boring work, they use computers to do it (faster) for them.

My computer is not altering its basic software as I type. Nothing I type in this text box will alter my operating system.

Anyway - what do you even mean by "database." Can you say what words you use to refer to actual DNA? And to an actual print-out of a human's DNA? I would say "DNA" and "DNA results." The ISL said they were 'lucky' to find enough complete DNA on the use point of the sheath (a common place to find complete DNA) that they could run several swabs. Each gave a complete human genome - they just didn't know who it was.

So are you claiming the lab altered the RECORD of the DNA (for what reason? why do you think that? I understand general mistrust of science, if that's your reason, but of course, I can't really have hope for a sensible conversation about it, if that's your plan).

OR, are you thinking that they altered the DNA itself? (Which I am telling you is not possible for the lab to do). Genetic engineering is a completely different thing. And since we now know that the DNA actually DID match Bryan Kohberger, we also know that from start to finish, the DNA was complete; it was compared to many other people's DNA at Othram (match game) and got matches. One of the closer matches (all partial of course, as BK wasn't in that database as a known individual) was a Kohberger, which immediately rang a bell to the investigators, as he was already on their radar.

Then they got Dad's DNA, ran it against the BK DNA (already run by ISL; with its record being entirely digital - NO ONE can actually see DNA with their eyes - except using electronic microscopes or similar, but even then, only biochemical analysis shows the SNP's). Dad's DNA matched at about 50% (proof of paternity - no one else in your world or mine is going to match you at 50% except your parents; keep in mind that these types of matches are ultimate run using the entire long set of graph points/letters (nucleotides) from the original DNA.

Database software. It compares. It does not alter. It wouldn't work (at all) if it altered and Othram and Gedmatch and 23andme and Ancestry would all go out of business. It just proves a READER that a human can use to understand - unlike the REAL results, which take years and years and teams of scientists to begin to ponder.

Here is one graphical representation of those base pairs (SNPs), and this is how it looks when we do it in the lab (this is a common staining technique under electron micrscope and represents a FRACTION of a person's actual DNA):

View attachment 458239

That's a stock photo from Getty. As you can see, you can't make heads or tails of it. Nor can I. This is just one computer-assisted way of lining up the original DNA results. But to actually multiple this picture by 1000 and then expect a human to figure out comparisons is impossible. We can't do it.

And here's a picture of the results of a paternity test:

View attachment 458240


Now, I know that those colors represent nucleotides, introns and extrons. So does the computer. In this case, we have selected a subset of the original data to compare (SNP's which are the only places that two individuals vary). Each of those colors represents a long list of even smaller data points (ACTG and more).

When we ask the computer to compare the above picture with someone else's paternity test, it will take us a method and quite some time to be able to see the differences (and similarities). It will not leap out at us. There's too many data points. Although, I've watched people work (and you can see Spencer Wells do it in his film Journey of Man, free on Youtube, at about the 45 minute mark - when he's in India). He's an expert in one chromosome. He's able to look at something like this (he prefers mathematical representations) and quickly see certain variations he's looking for. Page 216 of the below article shows how one single element in the chart above (purine) can be represented mathematically in relationship to other parts of DNA:


Keep reading to get to how to look at ONE gene mathematically (it's pages and pages - hopefully you can get a sense of what the computer is actually doing - it's doing math, quickly).

Here's what ONE gene (the gene that makes our blood red) looks like across species in its basic form:

Table 2. Listing of the bases of the first exon in the beta globin gene for the eight speciesmentioned. (Note: All the papers have used 90 bases for the rabbit exon 1 but it should be 92bases. Here we report the corrected sequence.)HUMAN (92 bases):ATGGTGCACCTGACTCCTGAGGAGAAGTCTGCCGTTACTGCCCTGTGGGGCAAGGTGAACGTGGATGAAGTTGGTGGTGAGGCCCTGGGCAG
GOAT (86 bases):ATGCTGACTGCTGAGGAGAAGGCTGCCGTCACCGGCTTCTGGGGCAAGGTGAAAGTGGATGAAGTTGGTGCTGAGGCCCTGGGCAG
OPOSSUM (92 bases):ATGGTGCACTTGACTTCTGAGGAGAAGAACTGCATCACTACCATCTGGTCTAAGGTGCAGGTTGACCAGACTGGTGGTGAGGCCCTTGCCAG
GALLUS (92 bases):ATGGTGCACTGGACTGCTGAGGAGAGGCAGCTCATCACCGGCCTCTGGGGCAAGGTCAATGTGGCCGAATGTGGGGCCGAAGCCCTGGCCAG
LEMUR (92 bases):ATGACTTTGCTGAGTGCTGAGGAGAATGCTCATGTCACCTCTCTGTGGGGCAAGGTGGATGTAGAGAAAGTTGGTGGCGAGGCCTTGGGCAG
MOUSE (92 bases):ATGGTGCACCTGACTGATGCTGAGAAGGCTGCTGTCTCTTGCCTGTGGGGAAAGGTGAACTCCGATGAAGTTGGTGGTGAGGCCCTGGGCAG
RABBIT (92 bases):ATGGTGCATCTGTCCAGTGAGGAGAAGTCTGCGGTCACTGCCCTGTGGGGCAAGGTGATTGTGGAAGAAGTTGGTGGTGAGGCCCTGGGCAG
RAT (92 bases):ATGGTGCACCTAACTGATGCTGAGAAGGCTACTGTTAGTGGCCTGTGGGGAAAGGTGAACCCTGATAATGTTGGCGCTGAGGCCCTGGGCAG

How quick are you at finding similarities there? I find the letters far easier to compare than the colors, but that's just me. Are you noticing all the similarities?

A computer that can mathematically deduce not only the comparisons but the actual biochemistry that these letters are indicating is the main tool for analysis. I noticed immediately that the lemur gene was closer to human than the mouse.

Analysis with a computer does not change the original data. It too knows that the first three letters of this particular gene (and the last three - they're call CODONS, and they are read by the body to produce every single thing that keeps you alive and makes your body what it is) are the same. Hmmm. So, across several species, there are commonalities in this gene. But what about the biochemical similarity? THere are multiple codons for each amino acid (and some don't code for amino acids at all). The computer knows this system and can immediately give the formula for the protein (and it better make hemoglobin!! And - it turns out, all of these genes DO make hemoglobin, through very similar biochemical processes).

50 years ago, we'd have only said, "it's hemoglobin, from a mammal." Now we can tell if it's Lemur hemoglobin or Human (and I have no clue what a Gallus is, maybe a type of rabbit?) Anyway, we now know that our closest living relatives of the group studied here is...well, can you figure it out? Or would you prefer that someone else ran this through an analyzer and computed the actual similarities and differences?

So that's what we're up to, in our labs. My work has to do with how to develop field methods for studying these things out in the real world. In humans. I don't study rats or lemurs. And I know quite a bit about the human genome, after 50 years in the field - BUT, I know almost nothing about any one particular SNP, because each of them would require me to take about a decade of the equivalent of post-doctoral study to be familiar enough to have a conversation with the people who are studying this.

Othram hires experts in genomic study PLUS computer experts and has managed to build one of the more robust SNP study databases (the datacomes from papers like the ones I'm quoting - that's the database). The incoming DNA from subjects is then compared to it.

(Sorry for the very long post - everyone here is so intelligent, but, we were born at different times - I know there is one younger WSer who already knows more about all this than I do, but may not have years of trying explain these methods both to grant writers, to students, to LE and to the general public). She may not have the patience either - I really feel it's important in True Crime for all of us to *try* and understand the science.
Click to expand...
Great post. It’s almost like people are forgetting that IGG is not a one way street that leads to a conclusion which is wholly reliant on LE’s interpretation of data.

We have a control to test against. It’s the original local ‘John Doe’ sample.

Which means the probabilities of the IGG process being junk and the output (Kohlbergers family) being wrong are somewhere in the neighborhood of 1 in 5 octillion.

Don’t quote me on that. I hate math. I’m just saying in order to believe one, you have to fly in the face of the other.

MOO
 
Jurisprudence said:
But, let's say for argument sake that it alters it completely. Devil's advocate, they built the tree on this "altered" profile, received an investigative lead, still did not arrest him but instead surveilled him, and witnessed what they did (him with the gloves at the store, cleaning the car, throwing his trash in the neighbor's bins). They took a sample from the family trash. It came back to statistically being the biological father of the perpetrator who left his DNA on the sheath at the scene. With that evidence, combined at that point with video evidence, cell phone data, DMV records, and more as evidenced in the PCA, they swore out the arrest warrant and retrieved his direct DNA either on-site in his kitchen or at the precinct (the warrant return of his person doesn't tell us where they were but I think that's irrelevant. We know it was on arrest). That direct DNA comes back as a match to the cs DNA.

How does your altered SNP profile help him? I think you are putting too much stock into "altered". I don't think they intend it the way you are receiving it.

jmo
Click to expand...
Actually, I think that what Thompson meant was not "altered" but that the SNP profile was sent from the lab as a dataset to Othram, then, in order to load the dataset into the Othram database, the dataset had to be translated into a format that could be read by Othram's database software. It is not unusual to do a translation when moving a dataset from one database to another, especially when more than one organization is involved. That is what I was trying to find out. I'm not saying anything happened to the dataset. Data Translation would make sense considering the SNP profile dataset was traveling to various different organizations that probably use different software and hardware. And yes, it would affect the format of the file including length.
 
10ofRods said:
I assume it was the Defense who used the word "altered." What gets altered is our perspective, just like using a microscope vs a telescope. Not the actual nature of the physical evidence or universe.
Click to expand...

I went back and watched. It kind of started when AT said that she believed there was missing information from the "private" (and we all know which one) lab's profile. Then she said that she had received in discover a SNP profile that she believed came from "private lab" and that it is not in the "general format that those come out in." She said it was not a FASTQ file but a different kind of format/file, and she said that it was a much smaller data file than the FBI's data file. She emphasized that she thinks the private lab's file/profile that was provided to the defense was "missing something."

The the judge said something like "Oh, I thought there was only one SNP profile and now I'm hearing there are two?"

quoting as best as I can hear from the not so awesome audio:

Bill Thompson said, "That's correct, your honor and the difference is that the one
from the private lab is the initial profile the private lab analyst created from the DNA portion of the DNA sample. Then what was transferred to the FBI was after that
original SNP had been uploaded into a database, access to the database was transferred. And it's our understanding that once a SNP is put into one of these databases, the database adjusts the SNP to fit what their criteria are for them to do their searches. So there are there differences between the two, the entirety of the two SNPs."

He then asks Attorney Nye, since he is knowledgable in this area, to jump in and correct hi if he's not stating it clearly.

BT continues, "So in an excess of caution we wanted to make sure the defense had both-- had the original SNP that the private lab created, and also the SNP (that?) was transferred via the transferring of the database to the FBI."

He then says, "We are told, we understand, that like I just said the databases adopt (* I think he meant "adapt"*) the SNPs to their search *audioblip*, and so they're going to look slightly different."

Judge J then asked Attorney Nye if he wanted to add anything to that, and Nye said no, he thought he was very well said.

AT then requests 2 things (which I'll discuss out of order):
1) She wants Judge J to explain what he means by in camera review and wants to know what the judge is going to do if he needs help understanding the info. She wants to make sure that there's not an FBI agent there with Judge J as he's trying to understand all this stuff and asks for him to clarify what he would do if he needs external help.
2) She wants the Judge to make a list (not the content though) of everything he gets from the State that he will then be going through and using to make his decision and then give it to her. Because she wants to make sure Judge J has a full packet of information upon which to make his decision. She specifically mentions:
--things like the raw data snip profile from "private lab"
--the communications between private lab and the FB--not the content, just that you have those
--the statistical analysis that happen within the profiles when making decisions when the genetics genealogists make decisions about the family tree
--the bioinformatics for the micro array
--the court should include that the FBI complied with requirements in MyFamilyTree

Judge J makes no promises to provide her with this privileged log of content. He says he might, it just depends on how things go. He also says that if he brings in an independent expert, he will let both sides know.

Mr. Thompson then replies that he is concerned about the list of things AT mentioned because based on the prosecution's understanding, there are items that AT thinks that should be there but that won't be because either
1) those things can't be provided to the State because they aren't kept per FBI protocol or
2) the defense has a fundamental misunderstanding of the difference between SNP profiles and traditional DNA comparisons, and how that process works. "We start talking with statistics, and those sorts of things don't necessarily apply in the investigative geneology world, so I don't want there to be false expectations. I certainly don't want the court be misled that things of relevance exist that do not...I was just just reacting to the laundry list of things the state was reciting that based on what our experts
told us, don't necessarily exist in an investigative genealogy SNP analysis. They may exist in a traditional STR DNA analysis, and those are two separate and distinct processes. They do not overlap at all, so we want to make sure that there's no confusion because
some of the things have been submitted by the defense don't accurately talk about the distinction."

AT replied that she does not want the judge to think she was trying to mislead the court or that she is misunderstanding or making things up in that list. "A lot of that
  • comes from expectations based on reading Department of Justice policy that was included for the court with our briefing before the August 18th hearing. A lot of that information comes from the testimony of the two genetic genealogist experts that we put on and the
    information that they provided to the court at the hearing. These aren't made up things. We absolutely understand that the genetic genealogy investigative process is different than what the Idaho State Police lab does with STR profiles, but there's analysis that happens along the way in building out these family trees to identify somebody
    that was talked about in deposition."

    BT then replies, "I'm not suggesting that they are trying to mislead the court. It is our position that they have been provided with information that's not accurate and it's creating false expectations. And I don't want the court to be misled by that and to have expectations that are not accurate."

    I do hope that Judge J does get an independent authority on DNA and IGG analysis to come talk to him who can explain things in a similar fashion to 10 of Rods. IMHO, BT was a bit inaccurate in his word choices. And I think Judge J SHOULD be allowed to have someone from the Justice Dept AND the FBI come and verify what procedure is and how it is followed in actual practice.
 
10ofRods said:
I have answered it every way I know how. The original data is not altered. It's just there. The original files and the original DNA is still there (in this case, on swabs that are preserved).

You are missing the point. Nothing is altered. Whose word is that, even? Did a geneticists say it? No, they did not.

If I draw a picture of you, I do not alter you. You are still there. Same with PROFILES. They are snapshots of the genome and each photographer does it differently. All are based on reality. Reality is not altered. DNA is not altered by the profiling process. In this case, it's also safely preserved in BK's body, as well as in the lab replication/analysis. It has been COPIED but not altered (altering DNA is an incredibly difficult thing to do requiring advanced technologies beyond the reach of any forensic lab).

But labs like Parabon and Othram never see that actual real DNA. They don't have blood from people. They don't even have saliva (well, they can collect it - but 90 percent of what they do is based on people's results from Ancestry and 23andme - where you do pay money to have your saliva analyzed).

The database is not altered. What database do you even mean? If you get blood test results that show you are anemic, is that result a "database"? DNA is a long string of graphical points (or letters, used as abbreviations). It's like a giant book that can be read by knowledgeable geneticists who then use computer programs to aid their reading (because no one understands the whole genome - each of these groups we're talking about is only studying a tiny fraction of the data).

All they are doing is basically putting different lenses on a kind of mental microscope. When you look through a microscope, nothing is altered in the slide (if you follow scientific procedures).

Nothing in the data is altered. The DNA submitted to Othram for the match (Kohberger's dad's dna) is still precisely what it was before (both physically and in its representation).

I can put it a slightly different way. In order to analyze SNP's, we MUST use computers. Some locations in the genome have up to 6000 different alleles that could be in the slot (and more are being found continually). But all Othram or Gedmatch are trying to do is MATCH one person's DNA (without any regard to what is being matched) to another person's DNA. In order to make this fast and not take years to get results, they FOCUS their lens on 600,000 SNP's (that's a lot). These are individual base pairs in the genome.

The FBI lens FOCUSES on only 18 (producing way more matches - as intended, as their goal is to continually scrutinize already known felons to make sure they're not committing new crimes). They don't care if they have to go to some guy on probation and make him account for where he was. They like doing that. I may be wrong about the exact number (it used to be 16, then it was 18, I haven't checked this year).

You want a MACRO lens for the FBI person, and you want a MICRO lens for matching two different people. You're playing a card game. One game has 18 cards, the other has 600,000 cards. The total deck is much bigger. The deck never changes (except by mutations in sex cells - which happen every generations but are of no value whatsoever in making familial connections, because that's the one region of the individual's genome where they will likely match no one).

If you take a snapshot of what you seem to want to call a "database" (to me it's not a database if it's just data - it's a database when you have individual files - and those individual files in the database CANNOT be altered or rewritten - that's fraudulent, unscientific and further, it would only change the letters in one version of a file).

The REAL Kohberger genome is now in the possession of the State of Idaho in several forms. Someone looking at the letters/graphical points that represent that genome, on a computer, at the other side of the country, is NOT altering any data. Couldn't get their work done properly if they did. They are simply looking for matches, just like that kids' game. And since it's super tedious and boring work, they use computers to do it (faster) for them.

My computer is not altering its basic software as I type. Nothing I type in this text box will alter my operating system.

Anyway - what do you even mean by "database." Can you say what words you use to refer to actual DNA? And to an actual print-out of a human's DNA? I would say "DNA" and "DNA results." The ISL said they were 'lucky' to find enough complete DNA on the use point of the sheath (a common place to find complete DNA) that they could run several swabs. Each gave a complete human genome - they just didn't know who it was.

So are you claiming the lab altered the RECORD of the DNA (for what reason? why do you think that? I understand general mistrust of science, if that's your reason, but of course, I can't really have hope for a sensible conversation about it, if that's your plan).

OR, are you thinking that they altered the DNA itself? (Which I am telling you is not possible for the lab to do). Genetic engineering is a completely different thing. And since we now know that the DNA actually DID match Bryan Kohberger, we also know that from start to finish, the DNA was complete; it was compared to many other people's DNA at Othram (match game) and got matches. One of the closer matches (all partial of course, as BK wasn't in that database as a known individual) was a Kohberger, which immediately rang a bell to the investigators, as he was already on their radar.

Then they got Dad's DNA, ran it against the BK DNA (already run by ISL; with its record being entirely digital - NO ONE can actually see DNA with their eyes - except using electronic microscopes or similar, but even then, only biochemical analysis shows the SNP's). Dad's DNA matched at about 50% (proof of paternity - no one else in your world or mine is going to match you at 50% except your parents; keep in mind that these types of matches are ultimate run using the entire long set of graph points/letters (nucleotides) from the original DNA.

Database software. It compares. It does not alter. It wouldn't work (at all) if it altered and Othram and Gedmatch and 23andme and Ancestry would all go out of business. It just proves a READER that a human can use to understand - unlike the REAL results, which take years and years and teams of scientists to begin to ponder.

Here is one graphical representation of those base pairs (SNPs), and this is how it looks when we do it in the lab (this is a common staining technique under electron micrscope and represents a FRACTION of a person's actual DNA):

View attachment 458239

That's a stock photo from Getty. As you can see, you can't make heads or tails of it. Nor can I. This is just one computer-assisted way of lining up the original DNA results. But to actually multiple this picture by 1000 and then expect a human to figure out comparisons is impossible. We can't do it.

And here's a picture of the results of a paternity test:

View attachment 458240


Now, I know that those colors represent nucleotides, introns and extrons. So does the computer. In this case, we have selected a subset of the original data to compare (SNP's which are the only places that two individuals vary). Each of those colors represents a long list of even smaller data points (ACTG and more).

When we ask the computer to compare the above picture with someone else's paternity test, it will take us a method and quite some time to be able to see the differences (and similarities). It will not leap out at us. There's too many data points. Although, I've watched people work (and you can see Spencer Wells do it in his film Journey of Man, free on Youtube, at about the 45 minute mark - when he's in India). He's an expert in one chromosome. He's able to look at something like this (he prefers mathematical representations) and quickly see certain variations he's looking for. Page 216 of the below article shows how one single element in the chart above (purine) can be represented mathematically in relationship to other parts of DNA:


Keep reading to get to how to look at ONE gene mathematically (it's pages and pages - hopefully you can get a sense of what the computer is actually doing - it's doing math, quickly).

Here's what ONE gene (the gene that makes our blood red) looks like across species in its basic form:

Table 2. Listing of the bases of the first exon in the beta globin gene for the eight speciesmentioned. (Note: All the papers have used 90 bases for the rabbit exon 1 but it should be 92bases. Here we report the corrected sequence.)HUMAN (92 bases):ATGGTGCACCTGACTCCTGAGGAGAAGTCTGCCGTTACTGCCCTGTGGGGCAAGGTGAACGTGGATGAAGTTGGTGGTGAGGCCCTGGGCAG
GOAT (86 bases):ATGCTGACTGCTGAGGAGAAGGCTGCCGTCACCGGCTTCTGGGGCAAGGTGAAAGTGGATGAAGTTGGTGCTGAGGCCCTGGGCAG
OPOSSUM (92 bases):ATGGTGCACTTGACTTCTGAGGAGAAGAACTGCATCACTACCATCTGGTCTAAGGTGCAGGTTGACCAGACTGGTGGTGAGGCCCTTGCCAG
GALLUS (92 bases):ATGGTGCACTGGACTGCTGAGGAGAGGCAGCTCATCACCGGCCTCTGGGGCAAGGTCAATGTGGCCGAATGTGGGGCCGAAGCCCTGGCCAG
LEMUR (92 bases):ATGACTTTGCTGAGTGCTGAGGAGAATGCTCATGTCACCTCTCTGTGGGGCAAGGTGGATGTAGAGAAAGTTGGTGGCGAGGCCTTGGGCAG
MOUSE (92 bases):ATGGTGCACCTGACTGATGCTGAGAAGGCTGCTGTCTCTTGCCTGTGGGGAAAGGTGAACTCCGATGAAGTTGGTGGTGAGGCCCTGGGCAG
RABBIT (92 bases):ATGGTGCATCTGTCCAGTGAGGAGAAGTCTGCGGTCACTGCCCTGTGGGGCAAGGTGATTGTGGAAGAAGTTGGTGGTGAGGCCCTGGGCAG
RAT (92 bases):ATGGTGCACCTAACTGATGCTGAGAAGGCTACTGTTAGTGGCCTGTGGGGAAAGGTGAACCCTGATAATGTTGGCGCTGAGGCCCTGGGCAG

How quick are you at finding similarities there? I find the letters far easier to compare than the colors, but that's just me. Are you noticing all the similarities?

A computer that can mathematically deduce not only the comparisons but the actual biochemistry that these letters are indicating is the main tool for analysis. I noticed immediately that the lemur gene was closer to human than the mouse.

Analysis with a computer does not change the original data. It too knows that the first three letters of this particular gene (and the last three - they're call CODONS, and they are read by the body to produce every single thing that keeps you alive and makes your body what it is) are the same. Hmmm. So, across several species, there are commonalities in this gene. But what about the biochemical similarity? THere are multiple codons for each amino acid (and some don't code for amino acids at all). The computer knows this system and can immediately give the formula for the protein (and it better make hemoglobin!! And - it turns out, all of these genes DO make hemoglobin, through very similar biochemical processes).

50 years ago, we'd have only said, "it's hemoglobin, from a mammal." Now we can tell if it's Lemur hemoglobin or Human (and I have no clue what a Gallus is, maybe a type of rabbit?) Anyway, we now know that our closest living relatives of the group studied here is...well, can you figure it out? Or would you prefer that someone else ran this through an analyzer and computed the actual similarities and differences?

So that's what we're up to, in our labs. My work has to do with how to develop field methods for studying these things out in the real world. In humans. I don't study rats or lemurs. And I know quite a bit about the human genome, after 50 years in the field - BUT, I know almost nothing about any one particular SNP, because each of them would require me to take about a decade of the equivalent of post-doctoral study to be familiar enough to have a conversation with the people who are studying this.

Othram hires experts in genomic study PLUS computer experts and has managed to build one of the more robust SNP study databases (the datacomes from papers like the ones I'm quoting - that's the database). The incoming DNA from subjects is then compared to it.

(Sorry for the very long post - everyone here is so intelligent, but, we were born at different times - I know there is one younger WSer who already knows more about all this than I do, but may not have years of trying explain these methods both to grant writers, to students, to LE and to the general public). She may not have the patience either - I really feel it's important in True Crime for all of us to *try* and understand the science.
Click to expand...
Bravo!
 
Status
Not open for further replies.

DNASolves

Who is DeKalb County John Doe (2022)?

Members online

... and 125 more.
Total: 2,246 (members: 126, guests: 2,120)

Online statistics

Members online
126
Guests online
2,120
Total visitors
2,246
Totals may include hidden visitors.

Forum statistics

Threads
602,256
Messages
18,137,664
Members
231,282
Latest member
omnia
Back
Top