tas8831
Well-Known Member
It has been claimed/implied by a creationist that the type of DNA analyses used in Courts of Law are more stringent and rigorous than those used in evolutionary biology - that such tests (those used in court) have been tested and are accurate. It has been indicated by this creationist that, in fact, these tests used in court are far better than those used in actual molecular biological analyses of phylogeny because they do not employ algorithms to assess the data, and further, that in courts of law, "genomes are compared side by side, loci by loci, not by matching by algorithms."
I will show unequivocally that this creationist's position is premised entirely on abject ignorance of the techniques employed; this creationist is uninformed and uneducated on these topics and has no business making such assertions on these and related topics.
1. DNA analysis techniques used in court.
For this segment I will rely primarily on the National Institute of Justice's "DNA Evidence Basics" page and the relevant links on that page, as well as the book "DNA Technology in Forensic Science". Any quotes used will come from one or the other of these sources.
There are generally 5 types of DNA analysis/analytical techniques employed by courts of law. Which type is used depends on a number of factors - cost, questions being asked, condition of biological material, etc.
They are:
Restriction Fragment Length Polymorphism (RFLP), Polymerase Chain Reaction (PCR), Short Tandem Repeats (STR, aka microsatellite), Y-Chromosome, and Mitochondrial DNA (mtDNA). The last 2 rely on analyses of specific markers found in these sources, so they basically fall under STR/RFLP anyway.
A brief explanation of each technique follows:
1. PCR - this technique can be used in and of itself, or in generating additional ‘raw material’ on which other analyses can be used. It makes copies of a target sequence, the end points of which are determined by the use of short DNA sequences called primers. In some cases, the length of the fragment produced by using the same pairs of primers can be used in a similar fashion to what is described below for RFLP. By definition, this does not use the entire genome.
2. RFLP – this technique uses bacterial restriction enzymes (enzymes that recognize specific DNA sequences, bind to them, and ‘cut’ the DNA at that point) to cut larger DNA fragments (or whole genomes) into smaller fragments. These smaller fragments are then run through a gel material using an electric current, and the fragments are separated by their length. Fragments of differing lengths, indicating the presence of indels, are the ‘length polymorphisms’ referred to in the RFLP moniker. These polymorphic fragments are heritable (or, rather, the DNA sequences that produce them are), so individuals sharing a certain number of them do so via ‘common ancestry’, so to speak. This technique may use an entire genome as raw material, but it is only the lengths of a set amount of fragments that are considered in the analysis – so, this is not the whole genome being compared ‘loci by loci.’.
3. STR (aka microsatellite) – In many areas of the genome, there are regions that are comprised of short tandem repeats – for example, 10s or hundreds of copies of TA (i.e., TATATATATATA…). Such areas are prone to polymorphisms, which again are heritable. The FBI has established a 13-locus strategy for DNA forensics (“For example, the likelihood that any two individuals (except identical twins) will have the same 13-loci DNA profile can be as high as 1 in 1 billion or greater.”). STRs are not the whole genome.
4. Y-chromosome – this only works with males, of course, so is often used in rape cases. It also relies on a set of markers (specific DNA sequences) on the Y chromosome (not the whole genome).
5. mtDNA – More useful in, say, identifying old remains in which the nuclear DNA has decayed sufficiently that RFLP or other analyses are not likely to work well. Regardless of why it is used, it, too relies on specific markers, not the whole mtGenome.
As to whether or not “algorithms” are used in these analyses in courts of law, a few link clicks from the NIJ site linked above shows us things like:
RFLP and PCR forensic analyses do not necessarily require software for analysis – it is just a matter of comparing band sizes in a gel (not whole genomes, not ‘loci by loci’), but it is probably the case that some analytical software is used, especially for searching databases and the like.
Just looking at these main techniques used by courts of law for doing various DNA analyses negates, 100%, the creationist’s claim that whole genomes are compared “side by side, loci by loci”.
2. As to how phylogentic analyses are done…*
Regarding phylogenetic analyses, PCR is a staple component.
RFLP - As I have previously documented, RFLP analyses have been used to assess Primate phylogeny, and the results were congruent with other DNA-based analyses (see, for example, this).
Strike 1.
STR – these have been used to assess modern human phylogeny:
“Reconstructing recent human phylogenies with forensic STR loci: A statistical approach”
Reconstructing recent human phylogenies with forensic STR loci: A statistical approach
as well as primate phylogeny:
“Microsatellite polymorphisms reveal phylogenetic relationships in primates”
Microsatellite polymorphisms reveal phylogenetic relationships in primates
Strike 2.
Y-chromosome – “A Molecular Phylogeny of Living Primates”
From the abstract:
“Gene partitions were analyzed separately, as well as combined, for genome comparison and phylogenetic reconstruction. Six gene partitions were created, corresponding to X-chromosome, Y-chromosome, autosome, intron, exon and UTR segments.”
A Molecular Phylogeny of Living Primates
Strike 3.
mtDNA – “Primate phylogenetic relationships and divergence dates inferred from complete mitochondrial genomes”
Primate phylogenetic relationships and divergence dates inferred from complete mitochondrial genomes
It should be noted that the human mtGenome is only on the order of 16kb
Strike 4.
3. Regarding the use of ‘matching algorithms’, it is hard to know what the creationist is actually referring to. It is the case that in his related rants, he brings up claims apparently made by disgraced hack Jeff Tomkins regarding his use of BLAST in assessing % sequence identity in humans and chimps (despite the fact that what Tomkins described, as relayed by the creationist in question, is not how such numbers are typically produced). The initial assessments of the % identity between humans and chimps was done using the entire single-copy genome of each in 1984, and while the analysis of those data was criticized by many, the values obtained were in the 90+%, so no ‘random matching’ involved. When the chimpanzee genome paper came out, there, too, a large proportion of both genomes were used, but not via ‘random matching’:
Best reciprocal nucleotide-level alignments of the chimpanzee and human genomes cover ∼2.4 gigabases (Gb) of high-quality sequence, including 89 Mb from chromosome X and 7.5 Mb from chromosome Y….
Genome-wide rates. We calculate the genome-wide nucleotide divergence between human and chimpanzee to be 1.23%, confirming recent results from more limited studies12,33,34. The differences between one copy of the human genome and one copy of the chimpanzee genome include both the sites of fixed divergence between the species and some polymorphic sites within each species….
No ’random matching’ – that is not how one gets an actual alignment:
There is a lot more after that quote, but it is pretty technical. It is from the Supplementary Data for the chimp genome paper.
**Please be aware, unlike some others, that BLAST is actually a collection of different programs. BLASTN, what Tomkins used, is a ‘random’ search program. BLASTZ is an alignment program – it does not randomly search for sequence identity
TAKE HOME MESSAGE:
When the creationist declared that in courts of law, “genomes are compared side by side, loci by loci, not by matching by algorithms”, he was either amazingly uninformed, was bluffing, or was engaging in mendacity. His untoward implications were dishonest, fallacious, and done out of either ignorance or malice (or both).
Of note, I came across an interesting discussion regarding Tomkins' use of BLASTN in his sad attempts to make it look like humans and chimps are not related (of note, he refuses to apply his same techniques to 2 random humans, or to pairs of creatures claimed by creationists to be from the same ‘Kind’). A few tidbits, in no particular order:
Lesson – best to take the proclamations of Tomkins with a grain of salt, and the second hand reporting of his claims by internet creationist types with 2 grains...
*one could just go here for the basics...
I will show unequivocally that this creationist's position is premised entirely on abject ignorance of the techniques employed; this creationist is uninformed and uneducated on these topics and has no business making such assertions on these and related topics.
1. DNA analysis techniques used in court.
For this segment I will rely primarily on the National Institute of Justice's "DNA Evidence Basics" page and the relevant links on that page, as well as the book "DNA Technology in Forensic Science". Any quotes used will come from one or the other of these sources.
There are generally 5 types of DNA analysis/analytical techniques employed by courts of law. Which type is used depends on a number of factors - cost, questions being asked, condition of biological material, etc.
They are:
Restriction Fragment Length Polymorphism (RFLP), Polymerase Chain Reaction (PCR), Short Tandem Repeats (STR, aka microsatellite), Y-Chromosome, and Mitochondrial DNA (mtDNA). The last 2 rely on analyses of specific markers found in these sources, so they basically fall under STR/RFLP anyway.
A brief explanation of each technique follows:
1. PCR - this technique can be used in and of itself, or in generating additional ‘raw material’ on which other analyses can be used. It makes copies of a target sequence, the end points of which are determined by the use of short DNA sequences called primers. In some cases, the length of the fragment produced by using the same pairs of primers can be used in a similar fashion to what is described below for RFLP. By definition, this does not use the entire genome.
2. RFLP – this technique uses bacterial restriction enzymes (enzymes that recognize specific DNA sequences, bind to them, and ‘cut’ the DNA at that point) to cut larger DNA fragments (or whole genomes) into smaller fragments. These smaller fragments are then run through a gel material using an electric current, and the fragments are separated by their length. Fragments of differing lengths, indicating the presence of indels, are the ‘length polymorphisms’ referred to in the RFLP moniker. These polymorphic fragments are heritable (or, rather, the DNA sequences that produce them are), so individuals sharing a certain number of them do so via ‘common ancestry’, so to speak. This technique may use an entire genome as raw material, but it is only the lengths of a set amount of fragments that are considered in the analysis – so, this is not the whole genome being compared ‘loci by loci.’.
3. STR (aka microsatellite) – In many areas of the genome, there are regions that are comprised of short tandem repeats – for example, 10s or hundreds of copies of TA (i.e., TATATATATATA…). Such areas are prone to polymorphisms, which again are heritable. The FBI has established a 13-locus strategy for DNA forensics (“For example, the likelihood that any two individuals (except identical twins) will have the same 13-loci DNA profile can be as high as 1 in 1 billion or greater.”). STRs are not the whole genome.
4. Y-chromosome – this only works with males, of course, so is often used in rape cases. It also relies on a set of markers (specific DNA sequences) on the Y chromosome (not the whole genome).
5. mtDNA – More useful in, say, identifying old remains in which the nuclear DNA has decayed sufficiently that RFLP or other analyses are not likely to work well. Regardless of why it is used, it, too relies on specific markers, not the whole mtGenome.
As to whether or not “algorithms” are used in these analyses in courts of law, a few link clicks from the NIJ site linked above shows us things like:
STR (Short Tandem Repeat) Data Analysis and Interpretation Software. Learn the basics of data analysis software, become familiar with the purpose of GeneScan® and Genotyper® software, learn the difference between GeneScan® and Genotyper® software and GeneMapper ID® software, and become aware of the unique features of GeneMapper ID® software, and understand FMBIO® Analysis software and STaRCallTM software as related to GeneScan® and Genotyper® software.
RFLP and PCR forensic analyses do not necessarily require software for analysis – it is just a matter of comparing band sizes in a gel (not whole genomes, not ‘loci by loci’), but it is probably the case that some analytical software is used, especially for searching databases and the like.
Just looking at these main techniques used by courts of law for doing various DNA analyses negates, 100%, the creationist’s claim that whole genomes are compared “side by side, loci by loci”.
2. As to how phylogentic analyses are done…*
Regarding phylogenetic analyses, PCR is a staple component.
RFLP - As I have previously documented, RFLP analyses have been used to assess Primate phylogeny, and the results were congruent with other DNA-based analyses (see, for example, this).
Strike 1.
STR – these have been used to assess modern human phylogeny:
“Reconstructing recent human phylogenies with forensic STR loci: A statistical approach”
Reconstructing recent human phylogenies with forensic STR loci: A statistical approach
as well as primate phylogeny:
“Microsatellite polymorphisms reveal phylogenetic relationships in primates”
Microsatellite polymorphisms reveal phylogenetic relationships in primates
Strike 2.
Y-chromosome – “A Molecular Phylogeny of Living Primates”
From the abstract:
“Gene partitions were analyzed separately, as well as combined, for genome comparison and phylogenetic reconstruction. Six gene partitions were created, corresponding to X-chromosome, Y-chromosome, autosome, intron, exon and UTR segments.”
A Molecular Phylogeny of Living Primates
Strike 3.
mtDNA – “Primate phylogenetic relationships and divergence dates inferred from complete mitochondrial genomes”
Primate phylogenetic relationships and divergence dates inferred from complete mitochondrial genomes
It should be noted that the human mtGenome is only on the order of 16kb
Strike 4.
3. Regarding the use of ‘matching algorithms’, it is hard to know what the creationist is actually referring to. It is the case that in his related rants, he brings up claims apparently made by disgraced hack Jeff Tomkins regarding his use of BLAST in assessing % sequence identity in humans and chimps (despite the fact that what Tomkins described, as relayed by the creationist in question, is not how such numbers are typically produced). The initial assessments of the % identity between humans and chimps was done using the entire single-copy genome of each in 1984, and while the analysis of those data was criticized by many, the values obtained were in the 90+%, so no ‘random matching’ involved. When the chimpanzee genome paper came out, there, too, a large proportion of both genomes were used, but not via ‘random matching’:
Best reciprocal nucleotide-level alignments of the chimpanzee and human genomes cover ∼2.4 gigabases (Gb) of high-quality sequence, including 89 Mb from chromosome X and 7.5 Mb from chromosome Y….
Genome-wide rates. We calculate the genome-wide nucleotide divergence between human and chimpanzee to be 1.23%, confirming recent results from more limited studies12,33,34. The differences between one copy of the human genome and one copy of the chimpanzee genome include both the sites of fixed divergence between the species and some polymorphic sites within each species….
No ’random matching’ – that is not how one gets an actual alignment:
Genome Alignment. We evaluated three independent methods for aligning chimpanzee contigs to human genome sequence. The first utilized BLASTZ (Schwartz 2003) to align and score non-repetitive chimpanzee regions against repeat-masked human sequence and BLAT (Kent 2002) to process the more repetitive regions. Alignment chains differentiated between orthologous and paralogous alignments (Kent 2003) and only “reciprocal best” alignments were retained in the alignment set.
There is a lot more after that quote, but it is pretty technical. It is from the Supplementary Data for the chimp genome paper.
**Please be aware, unlike some others, that BLAST is actually a collection of different programs. BLASTN, what Tomkins used, is a ‘random’ search program. BLASTZ is an alignment program – it does not randomly search for sequence identity
TAKE HOME MESSAGE:
When the creationist declared that in courts of law, “genomes are compared side by side, loci by loci, not by matching by algorithms”, he was either amazingly uninformed, was bluffing, or was engaging in mendacity. His untoward implications were dishonest, fallacious, and done out of either ignorance or malice (or both).
Of note, I came across an interesting discussion regarding Tomkins' use of BLASTN in his sad attempts to make it look like humans and chimps are not related (of note, he refuses to apply his same techniques to 2 random humans, or to pairs of creatures claimed by creationists to be from the same ‘Kind’). A few tidbits, in no particular order:
“If you mean the use of “-max_hsps 1” I’m not sure it’s an error, as Tomkins goes out of his way to use it and has done so repeatedly.
For those who don’t know, setting max_hsps to 1 means that the BLAST will only return a single alignment result per query sequence. “
“As mentioned by others, he used an ungapped analysis which means that a 300 bp comparison that differs by a single indel in the middle of the sequence would be counted as ~75% similar even though it differs by a single base.”
“It is interesting that Tomkins’ previous methods actually agreed with the consensus once the bugs were removed and gaps were accounted for properly. It would appear that Tomkins is in search of a method that will give him the results he wants.”
For those who don’t know, setting max_hsps to 1 means that the BLAST will only return a single alignment result per query sequence. “
“As mentioned by others, he used an ungapped analysis which means that a 300 bp comparison that differs by a single indel in the middle of the sequence would be counted as ~75% similar even though it differs by a single base.”
“It is interesting that Tomkins’ previous methods actually agreed with the consensus once the bugs were removed and gaps were accounted for properly. It would appear that Tomkins is in search of a method that will give him the results he wants.”
Lesson – best to take the proclamations of Tomkins with a grain of salt, and the second hand reporting of his claims by internet creationist types with 2 grains...
*one could just go here for the basics...
Last edited:
