That mutations happen?
Are you disputing that?
Of course not...indeed they do!
That germline mutations can be passed on the offspring?
Are you disputing that?
That’s how we inherit them...human babies inherit GERMLINE mutations from their human parents.
That mutations that are passed on can accumulate in descendants?
Are you disputing that?
Not at all.
That these patterns of mutations can indicated kinship?
Are you disputing that?
Germline mutations do indicate kinship! Only there are no germline mutations between apes and humans!
ASSUMING that which you want to believe.
I believe that I have, on several occasions, provided three citations indicating the reliability of methods used to assess hypotheses of phylogeny. And on the occasions that you have deigned to respond to them, you, like all other creationists thus far, simply dismissed them because the 'mice were still mice', totally ignoring (on purpose) the reason those citations were posted AND what they actually indicate.
In fact there are no GERMLINE mutations between most types of apes, though some apes do have similar mutations with other apes, and even with some monkeys, and humans also share some of these similar mutations but they are not inherited from one another.
What is your EXPLANATION as to why we can assess mutational patterns within groups, but not outside of those groups? Do you not accept that insertions or deletions can occur in the germline? If not, why not?
At what point do these analyses become suspect, in your oh so educated and informed opinion?
That is, when looking at germline mutations and inheritance, what, exactly, prevents these analyses from being used to compare different taxa?
All your years in Biotech, and you are not familiar with the act of insertion? Transposition?
Sure I am but the fact is mostly all alleged “insertions” have never actually been shown to have been inserted.
That is a fact?
Are you saying that all the things considered to be indels in human genomes are not really indels?
Or just when comparing different taxa?
Thus you negate your earlier statement - sure you understand insertions, but no, the things actual geneticists and the like see as insertions are not because they have not been shown to be?
You must know that any 2 human genomes differ by many millions of bps, yes? And that this includes one person possibly having a larger genome than another, yes?
If you accept that insertions occur, how, EXACTLY does pshun2404 determine whether or not what we see in a DNA sequence is or is not the result of insertion/deletion?
Funny - hundreds of thousands of degreed professional geneticists, population geneticists, evolutionary biologists, etc. can do this, yet pshun2404 the autodidact claims otherwise.
Please EXPLAIN your position, and support it with something beyond a couple of sentences or analogies or omissions.
They are simply there in one unique creature and not there in another unique creature.
Just so stories are so precious.
But not evidence, rationale, or explanation.
Show me genetic sequences that just happen to look a certain way, as you indicate.
Here is a great source for millions of sequences for you to use as examples:
Home - Gene - NCBI
Surely, you are familiar with this site?
In these cases (most idels fall into this category) it is just as likely these are just the normal parts of that creature’s genome. I have been through this before in another thread.
Nice assertion.
Yes, and I am sure that in that other thread you provided all kinds of evidence that supports your unorthodox take on the existence of things that are, you know, actually called insertions and deletions for a reason that apparently was overlooked by every other person researching genetics - perhaps you support it with a Blum quote, or the way you supported your 'SAME GENES' thing?
And I never referred to both patterns being in two humans...I said when we compare the genomes of two different creatures the program automatically creates these gaps (in order the “ALIGN” similar sequences) which is then interpreted as insertions or deletions (whichever is convenient).
I don't care what you wrote. If you think that when comparing two humans these things do not come up, then you have no idea what you are talking about.
I see indels when comparing two chimps or two macaques - as well as 2 humans.
In the creatures themselves these gaps DO NOT EXIST.
Thank you Dr.Gish.
OBVIOUSLY there are no 'gaps' in a genome.
Also, there are no petri dishes in nature, so surely you reject all information gleaned from things grown in petri dishes?
Here is how it works -
"Actually, for about 1/4 of your genome, you and your sibling are like identical twins, i.e. you have the same two parental copies of the DNA. Insertions and deletions (nicknamed indels) of up to about 100 bases are harder to enumerate but an order of magnitude of 1 million per genome is observed, about 3000 of them in coding regions (so an underrepresentation of about half an order of magnitude). Larger variations of longer stretches including copy number variations are in the tens of thousands per genome but because they are such long stretches their summed length might be longer than the number of bases in SNPs."
So, please explain to us how that was determined (yes, that particular source is using estimates, but it is based on published data). How is it that indels can be seen in human genomes when 'gaps aren't there' and nobody observed the insertions/deletion occur?
Below is the result of a pairwise comparison of 2 variants of the human KLF11 mRNA.
EMBOSS Water - Alignment
My goodness - a bunch of INDELS right there at the beginning!
7 of them!
Please explain how that can be!
Yes, the actual sequences are 'how they are' in the humans, but when comparing the sequences, those evil programs - designed to find similarity, according to you - insert GAPS where the sequences do not align.
By the way - most of the alignment programs I have used spit out %similarity scores just as an aside - it falls out of the actual comparisons.
Of course, if the program tells you 2 sequences are 94% identical, it also tells you - ready for this? - that they are 6% NOT identical.
Phylogeny programs are more interested in the 6% dissimilarity... if anything, they assume that the majority of the sequences will be identical.
The actual genome has these as differences and the program ignores these as contrary to its purpose of finding and establishing areas of similarity (that’s its purpose). Are you disputing that?
Yes, very much. I very much dispute your characterization of these software programs based on your need to dismiss their outcomes.
The programs do NOT ignore them!
Alignment programs are what FIND THEM (can also be done by eye - as alignments were actually done for many years).
The programs are not designed to 'find similarities' as you say (unless it is a program specifically told to do so), in fact, for the most part, they are 'designed' to find patterns of differences, wherein they insert gaps.
For how else to glean relevant information from a situation like this -
We have 2 sequences to compare:
1. ATCGGTCCCCGTCACACAAATTTGTCGTCCCGTAAGTGTGTGT
2. ATCCCCCGTCACACAAATTTGTCGTCCCGTAAGTGTGTGT
I use one of the biased, untrustworthy, anti-Jesus programs that you have declared exist solely to find similarities and ignore differences and I get:
1. ATCGGTCCCCGTCACACAAATTTGTCGTCCCGTAAGTGTGTGT
2. ATC- - -CCCCGTCACACAAATTTGTCGTCCCGTAAGTGTGTGT
And I conclude that an indel exists at this locus. I cannot tell whether it is an insertion or a deletion at this point for I have no other reference (you need 3 or more sequences to determine polarity), but I conclude that it is an indel. This means that, in this locus, these two sequences are 93% similar (remember, this is just an alignment), with a 3-bp indel.
Goodness no! GAPS! Gaps do not exist in real life! This is all a scam!
Pshun2404 looks at the two sequences:
1. ATCGGTCCCCGTCACACAAATTTGTCGTCCCGTAAGTGTGTGT
2. ATCCCCCGTCACACAAATTTGTCGTCCCGTAAGTGTGTGT
and concludes that this is just the way the Designer made them, and they are but 7% similar.
Let us use a non-DNA analogy to see which 'ideology' is more reasonable.
Say we have two lines of humans. One line has 100 dudes standing in line wearing red caps. The other has 100 dudes with the first 3 in line wearing black caps and the remaining 97 wearing red caps.
I say that the lines are 97% similar. Pshun2404 says they are 0% similar.
So, now we have the alignment. We can then test phylogenetic hypotheses (you need more than 2 sequences or taxa, so lets just consider that we do have more).
Depending on the program, indels are counted as either single mutational events or each bp difference is counted individually (not very realistic since an insertion or deletion is generally a 1 time event).
Say we have these 10 sequences:
1. ATCGGTCCCCGTCACACAAATTTGTCGTCCCGTAAGTGTGTGT
2. ATCCCCCGTCACACAAATTTGTCGTCCCGTAAGTGTGTGT
3. ATCCCCCGTCACACAAATTTGTCGTCCCGTAAGTGTGTGT
4. ATCCCCCGTCACACAAATTTGTCGTGTAAGTGTGTGT
5. ATCCCCCGTCACACAAATTTGTCGTGTAAGTGTGTGT
6. ATCCCCCGTCACACAAATTTGTCGTCCCGTAATTTGTGTGTGT
7. ATCCCCCGTCACACAAATTTGTCGTCCCGTAATTTGTGTGTGT
8. ATCCCCCGTCACACAAAGTCGTCCCGTAAGTGTGTGT
9. ATCCCCCGTCACACAAATTTGTCGTCCCGTAAGTGTGTGTTTT
10.ATCCCCCGTCACACAAATTTGTCGTCCCGTAAGTGTGTGTTTT
WE RUN THEM THROUGH OUR EVIL, BIASED, ASSUMPTION-LADEN ALIGNMENT PROGRAM AND WE GET:
1. ATCGGTCCCCGTCACACAAATTTGTCGTCCCGTAA- - -GTGTGTGT- - -
2. ATC- - -CCCCGTCACACAAATTTGTCGTCCCGTAA- - -GTGTGTGT- - -
3. ATC- - -CCCCGTCACACAAATTTGTCGTCCCGTAA- - -GTGTGTGT- - -
4. ATC- - -CCCCGTCACACAAATTTGTCGT- - -GTAA- - -GTGTGTGT- - -
5. ATC- - -CCCCGTCACACAAATTTGTCGT- - -GTAA- - -GTGTGTGT- - -
6. ATC- - -CCCCGTCACACAAATTTGTCGTCCCGTAATTTGTGTGTGT- - -
7. ATC- - -CCCCGTCACACAAATTTGTCGTCCCGTAATTTGTGTGTGT- - -
8. ATC- - -CCCCGTCACACAAA- - -GTCGTCCCGTAA- --GTGTGTGT- - -
9. ATC- - -CCCCGTCACACAAATTTGTCGTCCCGTAA- - -GTGTGTGTTTT
10.ATC- - -CCCCGTCACACAAATTTGTCGTCCCGTAA- - -GTGTGTGTTTT
I can see some interesting patterns emerge - for example, I would conclude that sequence 1 experienced an insertion of GGT based on the fact that it is the only sequence to possess it (it could also be that sequences 9-10 share a common ancestor that lost that GGT, and there would be ways to tell, but I don't want to get too far into the weeds with this off-the-cuff made-up example).
Pshun would have us believe that all we can say is this:
1. ATCGGTCCCCGTCACACAAATTTGTCGTCCCGTAAGTGTGTGT
2. ATCCCCCGTCACACAAATTTGTCGTCCCGTAAGTGTGTGT
3. ATCCCCCGTCACACAAATTTGTCGTCCCGTAAGTGTGTGT
4. ATCCCCCGTCACACAAATTTGTCGTGTAAGTGTGTGT
5. ATCCCCCGTCACACAAATTTGTCGTGTAAGTGTGTGT
6. ATCCCCCGTCACACAAATTTGTCGTCCCGTAATTTGTGTGTGT
7. ATCCCCCGTCACACAAATTTGTCGTCCCGTAATTTGTGTGTGT
8. ATCCCCCGTCACACAAAGTCGTCCCGTAAGTGTGTGT
9. ATCCCCCGTCACACAAATTTGTCGTCCCGTAAGTGTGTGTTTT
10.ATCCCCCGTCACACAAATTTGTCGTCCCGTAAGTGTGTGTTTT
Created as-is. Nothing to see here.
Too bad that the tested methods that I have referred to call such a conclusion nonsense.