first draft for autism markers

[mellowtigger]

First draft, but I just want to get something "finished" enough to post, because I have other tasks to work on this week and need to try getting this one out of my head for a short while.  Summary:  I have 4 autism markers, 2 uncertain, several normal markers, and many that remain untested or unexplained.

gene	address	normal	autism risk	me	reference
MET	rs1858830	GG	CC x 2.27, CG x 1.67	CC	PubMed
ADRB2	rs1042714	CC	GG x 1.33 - 1.60 (or even twice that)	GG	PubMed
GABRB2	rs2617503	CC or TT	CT x ?	CT	ScienceDirect
APC	rs1804197	AA	CC x ?	CC	PubMed
NPAS2	rs1811399	?	C (unspecified heterozygous risk)	AC	Nature
EN2	rs1861973	?	C (unspecified heterozygous risk)	CT	Nature

For many autism markers, I have the typical dna pattern instead.  There are a lot of markers that are not tested by 23andme.  There are many studies which mention specific SNP addresses but their summaries do not indicate which base pairs were involved (and I can't afford to purchase all of those articles for their details).

Some markers were tested, but I had results not indicated in any expected combination.  For example, I have AA at rs2710102 (CNTNAP2 gene) but expected values according to the article are CC, CT, and TT for Caucasian Europeans.  Similar shenanigans occur at rs1322784 (DISC1 gene).  What these differences are supposed to mean, I have no idea.  Are they:

1. bad gene chip results (disappointing),
2. contamination (also disappointing),
3. brand new mutation (exciting!), or
4. a data mixup in how the SNP addresses are reported (confusing)?

Apparently there's some kind of left-hand versus right-hand notation involved in naming the "rs#" addresses, so it could easily be a translation problem there.   Some texts, however, were just plain confusing, like rs1861972 and rs1861973 which seem to simultaneously associate for and against autism, unless I'm misreading that somehow.  I didn't get enough sleep last night, so maybe I should try again in a few days with a fresher mind.

I still have more references to check.  There are so many autism markers!  More to say in the second draft later, I guess.

Comments

[maradydd.livejournal.com]

This is really fascinating. I admit I'm a little out of my depth here in terms of answering your questions or even shedding much insight, but I am learning as a result of observing your process. Thank you for doing this! Reading these entries is like opening a present. :)

Since you're posting them publicly, would you mind if I passed the URL to your posts with the "gattaca" tag to the DIYbio mailing list? I think other people would also find them instructive, and others may be able to help shed light on things that are confusing about notation or whatever.

I'd like to know more about autism markers, as well. I won't have the money to spring for a 23andMe test anytime soon, but I was diagnosed with Asperger's syndrome in 2004 and am curious what markers I might have.

[mellowtigger]

I'm not a fan of the notion of intellectual property, so feel free to pass the information along anywhere that it might be useful. :)

[lordalfredhenry.livejournal.com]

Cool! I need to review myself.

rs1858830: GG normal
rs1042714: CC normal
rs2617503: CC normal
rs1804197: CC risk?
rs1811399: AC ? matches yours.
rs1861973: CC ? you got CT.

Did you get prometheaseZ to work with what I mentioned on directory structure or was there more work to do involved? That's really interesting. Now I want to do some reading of those articles. :)

[mellowtigger]

rs1858830: GG normal
rs1042714: CC normal
I had quite a few normal results too. I did not cross-reference all of those links, though, to add them to the table. Maybe on a 3rd or 4th draft of this table. :)

rs2617503: CC normal
rs1804197: CC risk?
The article summaries did not specify the risk level associated with the rogue nucleotide. If I could buy the articles, it probably has that detail in it.

rs1811399: AC ? matches yours.
rs1861973: CC ? you got CT.
The article here mentioned only the "C" allelle, so I'm sure that your CC result would qualify. I'm just not sure if my own "CT" conveys any risk or not. Again, more information in these article summaries would help a lot!

rs2710102
rs1322784
What were your results here? We can suspect translation problems if your results are also "out of line" with expected values.

Promethease program
No, I didn't have any luck getting the most recent program to run on my Vista 64-bit system at home. I did get the older version to work, but I still had to manually select a new output directory for my results file. It would not write to the default directory, for unknown reasons.

[lordalfredhenry.livejournal.com]

rs2710102: GG
rs1322784: GG

I usually score high on all the delayed speech development SNPs. IIRC, these were some suspect ones. I'm guessing GG is equivalent to CC. (reversed pair) I found more:
http://www3.interscience.wiley.com/journal/112224058/abstract?CRETRY=1&SRETRY=0


http://en.wikipedia.org/wiki/DNA#Base_pairing

[mellowtigger]

Ok, so it looks like it is a matter of "translation" then. Cytosine and thymine are expected at this location, according to the SNPedia entry, but you and I have guanine and adenine instead. Left/right confusion. But if all of our other addresses are appropriately labeled, then why would these two particular entries be different in the SNPedia? *frustration*

[khrysso.livejournal.com]

This exploration is all so interesting! I feel very lucky to be privy to it.