Personalized genetics?

It seems like all my genetics/biologists friends know about this, but whenever I mention it to my proteomics friends no one is familiar..and it's surprisingly relevant to human proteomics.

23andMe is a personalized genetics service that uses topnotch technology to sequence important sections of your DNA and give that data to you.  I've thought about it for years (In 2008 Time Magazine named it the invention of the year), but I was pretty sure it would end up invalidating me for health insurance cause of some pre-existing condition I didn't know about.  Thanks to the Affordable Care Act, I don't have to worry about pre-existing conditions anymore, so I ordered a kit and had it shipped to my parents in WV cause you can't get them delivered here in MD.

This service used to provide you with disease information, but they can't anymore due to some lawsuits from doctors or something, but they provide you with your RAW sequencing data and there are plenty of genomics tools out there.  I've been digging through mine big genetics file with the Interpretome software (article here, and open access!)  There is so much data.  But that isn't the real story.

The real story is the huge amount of genetic variation that we have.  In proteomics, I don't think we consider it a lot.  Seriously.  We take the peptide MS/MS data from every human sample and we compare it vs. Uniprot.  We're making the assumption that the proteins from my plasma are going to match the protein sequence of the proteins from the FASTA of the sequenced person in the UniProt file.  And for the most part, we're probably right.  Albumin is albumin.  It is pretty well conserved among all mammals.  But what about other proteins?  Just randomly selecting some data from my genomics data and a pathway that is pretty well annotated:

Parkinson's disease	4698412	AG
Parkinson's disease	9917256	GG
Parkinson's disease	356220	CC
Parkinson's disease	12726330	GG
Parkinson's disease	1296028	AG
Parkinson's disease	11026412	GG
Parkinson's disease	2839398	CC
Parkinson's disease	6430538	CT
Parkinson's disease	10877840	TT
Parkinson's disease	12456492	AA
Parkinson's disease	199515	CG
Parkinson's disease	11865038	CT
Parkinson's disease	2395163	TT
Parkinson's disease	11248060	CC
Parkinson's disease	356220	CC
Parkinson's disease	3129882	AG
Parkinson's disease	356219	AA
Parkinson's disease	11724635	AC
Parkinson's disease	1491942	CC
Parkinson's disease	199533	AG
Parkinson's disease	199533	AG
Parkinson's disease	17115100	GT
Parkinson's disease	1223271	AG
Parkinson's disease	6532197	AA
Parkinson's disease	2736990	AA
Parkinson's disease	393152	AG
Parkinson's disease	6812193	CT
Parkinson's disease	7077361	TT

Check this out!  Every place in this pathway is a place where we are known to be genetically different in the pathways we currently know lead to Parkinson's disease that are tested by this $100 kit.  Those letters at the end?  Those are the nucleic acids that vary between different human beings.  Look at the first one.  At that point in the DNA you can either have an A or a G.  If that is the third letter in a codon, this is probably not a problem:

The third letter shift from A to G rarely results in a different amino acid being put into place.  Look at isoleucine, though.  ATA means isoleucine.  ATG means methionine!  If it is the first in the codon, then it commonly means that your peptide sequence has a different amino acid in it than mine does.  And if I'm searching my peptides vs your protein sequence there is nothing for that spectra to match to and that spectra goes into the trash.

I find this 1) a little scary and 2) super interesting and 3) exciting, cause someone out there is going to solve this stuff and I can't wait to see how you do it!

I've been pretty philosophical here the last two days.  Its because I'm doing some hard technical stuff all day. Results are on the way, I promise!