No wonder I was having so much trouble getting a straight answer on my "Adding Information to the Genome" thread.
The reason you weren't getting an answer, or at least not one that satisfied you, was because you couldn't articulate a coherent concept of information that could actually be measured. People gave you lots of perfectly straight answers, you just didn't like the measures of information they were using. When you did give a specific measure or at least specific forms of mutation you were interested in, i.e., single nucleotide substitutions or duplications at various scales, people provided examples where they did have functional effects.
I can't understand why it's termed "junk DNA" when similar sequences are found inactive in other primates. You'd think they would differentiate between junk and an inactive gene
There is no way to distinguish between them in the absence of a reasonably sized coding region. The sequences don't have the sort of homologies we can use to identify functional genes in a pathway of common descent. There are lots of pseduogenic sequences in what is commonly thought of as 'junk DNA' that would be more easily identified as 'inactive genes'.
In fact, this doesn't make sense. Why do they believe the genes originated from junk DNA, rather than from "rearrangements of existing genes"? The fact that the sequence doesn't code in primates doesn't mean it's not an existing gene, does it?
The non-coding forms of the sequences aren't just found in the primates but also in other more distant species, which is why the most parsimonious conclusion is that they were newly generated in humans. The coding forms of these sequences currently appear unique to humans, so all the evidence is that these actually are novel genes. The lack of homology to other known proteins argues against an origin from the re-arrangement of existing coding sequences.
Highly frustrating that they don't know what these genes code for, other than the heightened activity of one in cases of leukemia.
Indeed. That and the lack of evidence for selection around the genes makes me wonder if maybe these particular sequences simply aren't functional, an interesting example of Kimura's theory. Although if the leukaemia associated gene was actually a causative factor you might think there would be some negative selection, that would probably depend on the type of leukaemia though since childhood leukaemias would be much more likely targets for selection. Lots of genes are non-specifically upregulated in cancers though so it may not have any causative role at all.
It is worth pointing out that they still don't actually know what the leukaemia associated one does, they just know that it's expression is upregulated in some cases of leukaemia.
TTFN,
WK
Edited by Wounded King, : No reason given.