Hacker News new | comments | show | ask | jobs | submit login

One interesting thing I learned by getting my genes sequenced is that provigil would apparently have no effect on me. This would probably account for why a lot of people think provigil/modafinil didn't do anything for them.

About 25% of people of caucasion decent are (A,A) at rs4680, which is in the COMT gene whose enzyme degrades dopamine/epinephrine. Modafinil probably works by raising these catecholamines (at least in part). I haven't read it but the study is here: http://www.ncbi.nlm.nih.gov/pubmed/19037200?dopt=Abstract

I also have A,A (Met) at rs4680. I wouldn't necessarily conclude that modafinil would have negligible effect on us based on that study. The study was narrow, it looked at recovery from sleep loss and looking more if dopamine signalling affected sleep recovery.

There's tons of interesting studies about rs4680 (http://www.snpedia.com/index.php/Rs4680). Although I think all the data is still to broad to make any conclusions.

I think I might give modafinil a go just to see what happens.

Hey, you prompted me to look at the paper :). My understanding is that the study primarily looked at cognitive performance during an extended period of wakefulness (40hrs). Two doses of Modafinil (100mg each) were given during this period (the manufacturer recommended dose appears to be 100-400mg).

Modafinil improved the cognitive performance of Val/Val's (G,G)s across most of the tests (memory tests, mood tests, "Vigor" tests, reaction rate tests, well-being, fatigue, are ones i remember). Met/Met's (A,A)s did not see any statistically significant improvement on most of the tests. One test where there was a small effect for Met/Met's was in subjective well-being.

The study also looked at sleep recovery after the prolonged wakefulness. All participants entered deep sleep faster than normal.

90% of Val/Val's correctly identified their group (placebo vs. drug). 66% of Met/Met's correctly identified their group.

There were 10 people in the Val/Val group and 12 in the Met/Met group. But the small number of participants should be accounted for in the p-tests, as far as I know.

However, like you, I'd still try it given the chance. Regardless, lets hear it for our naturally increased dopamine!

"I think I might give modafinil a go just to see what happens."

Has everyone on HN gone crazy for the day or something?

These threads come up two or three times a year. Usually, the people who indiscriminately encourage everyone to pop pills get voted way up.

I'm seein' robots.

If you think of all of HN as the single person in this comic: http://xkcd.com/242/

My friend is rs4680(A, A) and he finds modafinil quite effective for increasing concentration, memory and enthusiasm for work. It didn't seem to be a lot of help for situations of sleep deprivation though which is what the study you refereed to was about. As opposed to the author's description, it did tend to make him somewhat 'high' and 'wired' as well as causing a degree of muscle tension and difficulty in 'coming down' at the end of the day. Also doses of 50-100mg seemed optimal and 200mg was way too much. It is difficult to know if this different response is related to rs4680(A,A) or not.

I've just checked the "drug response" section in my 23andme.com data, but I don't see anything about provigil.

Searching for rs4680 did show a result though (I have the AA variant).

This means, like me, you have a Valine codon in place of a Methionine codon in your COMT gene at codon 158. This means the enzyme catechol-O-Methyltransferase which breaks down catecholamines like dopamine/norepinephrine has reduced activity and less stability, resulting in increased catecholamines in some areas of the brain.

I recommend downloading your raw data from 23andme and using the prometheus program from SNPedia. This is how I found out about this gene. This program is very raw but includes a lot more research and points you to the relevant papers.

Keep in mind, this information is all AFAIK, as I am just browsing this information casually.

> prometheus


(edit) Oh, it's a crippleware. They want "optional $2 payment" so that the program would run in 5 minutes instead of 3 hours. How cute.

I took some generic modafanil and it worked as I thought it would, though the traditional stimulants are clearly more effective. I got my 23andme results back the next month and ran it through SNPedia and saw I was a non-responder to Modafanil, but that mutation is definitely not an absolute determination.

Where does one go and how much does it cost to get your genes sequenced these days?

$400-$500 US at https://www.23andme.com/

They had a special deal for $99 when I got it, but that was very limited and they haven't done it since. Still I'd say if you hold out for a year that price will be halved. I generally recommend people to wait unless they're really interested in these things. The industry is still in its infancy. Cost is going down quickly and quality is going up just as fast.

Ah, interesting. It's not a full genome sequence, just a sequence of strategically important loci, I see.

Yes, you get about 570,000 single nucleotide polymorphisms (SNPs).

And as far as I know, the value of SNPs is debatable at best.

I used 23andme. I was lucky enough to be notified they were having a one-day sale a couple months ago. The sale price was $99 for the full reults, which is 80% off of their usual price of $499. The sale was on national DNA day.

There's at least a couple other similar companies out there, with more well known ones being deCODE and Navigenics. Recently these companies were getting in trouble with california health regulators for providing medical information without a doctor (I think that was the complaint, but I don't remember the exact reason). But the California complaint included 12 companies so there must be other companies as well.

23andme is one such service. I have GG at rs4680.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact