promethease 2013 0.1.156

This video demonstrates how to use Promethease as of July 2013. To begin I’m going to need a copy of my genome. If you don’t already have your DNA there
isn’t much that we can do for you. My DNA has already been analyzed by a
company named by visiting I’m taken to a page where I can login and download my information. I will
put in my password and download, not my data but the example data from Lilly Mendel. Now that I have her data on my computer I can
come back to . You might notice the use of HTTPS for improved security. The files
from many companies such as, FamilyTree DNA and
many many more are all understood. You can upload any of them. I’ll click on
‘upload from your computer’ and here in my downloads directory, the
same place I downloaded the Lilly Mendel genome I’ll choose that file. You’ll see a
progress bar as the upload is transferred. if you don’t see that
kind of a progress bar perhaps you are using a very old version of Internet Explorer. Updating to a
recent version of Chrome or Firefox is strongly suggested because other features later on may also
have problems it recognizes this as a 23andMe format file and tells me I
need to pay five dollars from click on this button and it remembers me at Amazon Notice that this is a different
user email. The account that you use for Amazon is
not necessarily the one that using for your 23andMe or Ancestry data. Because I’ve bought books and things previously, Amazon already
has my credit card and knows me. To continue just put in the captcha information to verify that I’m a
human. Confirm the payment and after about five seconds I’m taken
back to The website now recognizes the payment
as complete. I’m given the choice to pick an ethnicity it’s not essential that there be an
ethnicity close to yours in here. for many of us we’re a mixure or we just don’t fit any of these categories. Pick whatever’s
closest it doesn’t really matter. It does
influence the sorting order of some parts in the report,
because we know certain things are more common or more rare in these populations, but fundamentally
all the same information is in there and everything is is fundamentally the
same. Put in an email address and/or a nickname. When the report
is emailed to you at the end of the run
you’ll be able to have a record of it that way. For 45 days you can retrieve the file
from our servers and after that point your data is simply
gone. We retain your report for 45 days but the data that you upload to us, your
raw data, is deleted seconds after your report is generated. We want to preserve your
privacy as much as we possibly can. I think we do a very good job of that. If
you have more questions about our privacy policy the front page has a link at the bottom
with a note about privacy and some text to explain all what’s
going on. It takes about 30 seconds for promethease to recognize that you
have made your payment and that you’ve filled in
your forms. At this point the processing really begins. This will typically take about 45
minutes that may change in the future, either up
or down, as I modify how the server works. While you wait there’s a video down
here that you can watch. This video is meant to help you understand
how to read your report. “Today I’d like to demonstrate some of
the features from Promethease reports after 0.1.122…” Because that report video is now
increasingly out-of-date what is there now is this video that you’re watching. While we wait, I do want to point out one
aspect of this video that I think is interesting to note. If I back up to the
very beginning here you can see that we were working in version 0.1.126 in January 20, 2012, when I recorded that video, at that time there
were 14,200 genotypes annotated. In just a few
moments we’ll be seeing what a report looks like about 18 months months, later with the latest
version which is currently 0.1.156 and instead of 14,000 genotypes I believe there’s now about 17,000
genotypes that will appear so what you can see is that the
software ‘Promethease’ is continuously being improved. There have been 30 versions in the last 18 months. Also the
content in SNPedia, the number of SNPs for which we know some kind of
information because additional scientific literature
has been published new papers in PubMed continues to grow, so there is a a reason
to come back and run a new Promethease report at some frequency. Some people may want
it once a year. Some people may want is once a month even if you wait just a day or two,
usually a couple new SNPs will appear. So after just about five minutes of
runtime the previous report has been fully
generated. You’ve also received the email but I’m
going to just click on the download button and bring this report in. Currently
you’ll see that its a 32 megabyte file if you’re working with
Family Tree DNA, Ancestry or others each of those has a different collection
of SNPs on the chip that they test and the size of the reports will be
different and what content is there will be different. Now this part here is actually the most
complicated piece it seems. Windows makes a little bit too easy to
go into that zip file but in not actually opened up so I’m instead gonna say ‘show in folder’ and now here it says ‘report’. That is a compressed zip file and by right-clicking on the
report I can say ‘Extract all’ and under Windows
this gives me a way to actually open up all the files that are
inside this collection, because the Promethease
report is not a single file, and yet we have to make a single
download. If you’re working under Macintosh or other operating systems
your mechanism for unzipping is up to you, and I’m sure you can figure
that out. So now I’ve been taken into the report
folder that was just uncompressed. You’ll see there’s both
a ‘report’ HTML file, and directory of supporting data and by double clicking on the report I’m taking in to the current report and so here you see version 0.1.156 the date on which this was generated, July
23rd 2013 and that almost 18,000 genotypes, 17,809 are currently known. You can click on good news and see just
the ‘good news’, or you could scroll through here and see the the same for ‘bad news’ interestingly there is a section just on
medicines but I really want to begin with what I think
is the most important, most interesting, and most useful, the so-called “UI version 2 interactive
report”. this is a continuously updated software
I’m trying to put together to make it as easy as possible for me,
and for you, to explore all these 17,000+ genotypes of
information it’s a lot of information and there’s no
one single best way to represent so here you can see that there’s boxes
outlined in grey which are not necessarily good or bad such as gender I color hair color not things outlined in green which are considered good and things outlined in
red that can be considered that how high up something is in this
order is a functional a a parameter we call magnitude and you
can filter by it over here discuss it later in more detail
magnitudes the way that the community decides how important is something is a2x risk of Alzheimer’s more
important than 1.6 X risk of prostate cancer and other
cancers well so far it seems so but we only
think that by a little bit 3.5 verses 3.2 that I color is just a little bit
further down and for you the order of all these
things will be different some things won’t be known it may not be possible to
classify your eye color other features quite as well but you can
come down to the bottom of this and see some other kinds of things that we are
able to classify and so far only the first 10 are visible
on screen while seventeen thousand plus are still
off-screen so clicking 2x more will double the
number thats on-screen and I can continue to scroll through
here and explore deeper and deeper into my data what I want to point out is that there’s
just too much to really displayed all at one time but I can begin to say well there’s
certain topics i care about. maybe it’s CA and see are and now all the steps that mention
cancer are brought to the front and now it’s a
much smaller list now there’s only 800 of them and the pie chart has been
updated to reflect how many of the ones that are currently
on-screen are considered bad apparently sixteen of them 195 were
classified as good news and the majority 75 percent to them 618 total are still not yet classified as
clearly good or bad over you notice the bad ones in red tend
to be at the top good news is usually pretty boring
because you probably don’t have any superpowers hiding inside your data and so a lot of the good news is down at
the bottom or the bad news tends to be at the top and I i’ve seen some people get a little
worry about this because the first thing they see is the bad but if we should do all the places where
your normal you be a little bored so right now we’re searching for all
kinds of cancer probably go ahead and get a little bit more specific here and
pick just the category lung cancer we’re still searching and
filtering by the word cancer but instead we’ve had to this other
section and hear lots of different topics like longevity leukemia leprosy are all here don’t get where the word cancer because
we’re still filtering by so go ahead and get rid of it and now
you’ll see that there are six leprosy Associated slips which are not
yet classified as good or bad and one which is classified as
necessarily good supposed to these are just not yet known
but by looking at the frequency this is the frequency in the chosen
population the CPU Caucasian population we can see that 30
percent to people have this genotype if thirty percent to people have it it
certainly is unlikely to be particularly bad rare there are somewhere that’s the case
let’s go ahead and do a different search I will get with the category of lepers look at things that have already been
classified for yet and we get breast cancer or breast
size we focused on breast cancer is it some
more interesting topic for most people and here we can see the top one is a
higher cancer risk the things that you see in this grey box
are specific to your Gmail type specific to being assisi at this
location by clicking on board you get more
information about the position generally and so the information in this
box or by clicking on more info is the same
information that you would get by clicking not on the genotype but instead over
here on the Rs number Rs numbers are standard identifiers from
the NCBI the National Center for
Biotechnology at the NIH the National Institutes of
Health dot gov all these have been assigned
these names by a US government authority and by clicking on
the information here I can see all the information that we
have already collected about this position in the genome generally
there’s 8 p.m. ID number that’s a link of to
PubMed and a brief summary of it often just the
title love the paper the gene that it’s in the chromosome
frequencies position all kinds of other information are already classified here generally
newer papers are down at the bottom because they’re populated automatically
fun almost every night bases but sometimes the important summaries
were sitting at the top in this case that’s not really the case is a few other details I can point out
here on the right hand side we’ve classified the CC is bad in red with the magnitude of 2.5 and the one brief summary if higher
cancerous follows GG is considered good and no and you can look then at the frequency
of the different genotypes in different populations hears that CE you that we chose for Lily
mandel and some other ones by clicking on the
question mark you’ll get information about each of those codes but they were roughly correspond to
caucasian Chinese Japanese African african-american
southwest’s and a few other populations like
Gujarati indians and Mexican population and Tuscan
southern Italian along with the average fall these and
we’ll see is that for instance the CC genotype here
colored in brown is fairly rare in caucasians but
relatively common for the Chinese and Japanese populations so the overall frequency is about 20
percent whereas the blue GG is really quite
common occurring in about fifty percent of all caucasians at this point I’ll come actually know I
will I will click on one of these PubMed ID’s and see what we can to learn about it by
going off to PubMed now were at that NIH dot gov
address you can read the abstract of the paper
sometime see some figures and what you’re really lucky you can go
ahead and click on one of the links here and read the full absolute paper for
yourself and that’s exactly what we’ve done to go
ahead and fill in the information that you see so far but we’re very much reliance upon you
and people like to come in and help filling information
into sticky the we all learn more about our genome so we’ve seen a little bit how you can
do filtering I’m gonna go ahead and sort not by magnitude not by the things that are already
classified as being very important but by frequency frequency is a good way
to find things that are fairly rare in specific to you in fact I’ll even
take it off for breast cancer I’ll come back in just show all and now
we can see a particular snack here only .9 percent so just a little under
one percent of Caucasians have this a age in a tight apparently it’s
slightly increases the risk for prostate cancer by clicking the more I can see some of
the more information about it clicking and more information would
again take me into that patients that pedia and I could see that it’s not just rare
for caucasians that AA is rare for every population and
GG really is much more common so we’ll come back into
the UI to report you will getting sort by
frequency here’s this cheesecake but nobody has
yet classified this one as good or bad I think it’s pretty clear from this
section here the it should be back I’m gonna go ahead and click on AAA and
taking in not to the general page about this that
but the information specific to this genotype somebody has already filled in the
Summary slightly increased risk for prostate cancer by clicking on edit before I’m gonna go
ahead next to change to a few so that this becomes officially bad to
leave a brief summary hopefully you have something better to
put in that most cases new Mayfair not logged in have to go
ahead and do a capture to prove human with that’ll scroll to the bottom save and that’s it I’ve just made state PD a
little better and I made all the future for me yes
reports a little bit better so anybody else who shares this fairly
rare genotype will immediately see it classified into the bad categories so you get a sense of what you can do
here will point out that this one was notable because it had 15
references 15 papers were already linked back on that Rs number page in it’s the pedia the next one is also quite rare at a little less than one
percent there’s only a single reference if I
click on the IRS number will see that in this case there’s one paper relating to bone mineral density here she was and well that’s that’s interesting but I don’t
really want to think about the ones only have a single paper state PDFs to capture them because we
can’t know that it has 15 papers unless we first know that has one and two and
three but I’m gonna go ahead and click on plus where you could use the slider and say
you know what I only want things ever least 12 references if there’s 12 papers okay now I believe it and we still see the one we saw a moment
ago but a bunch was been dropped out as not yet particularly credible how we can
instead focus on just the ones that have more information but when I do this I’ve
got to ten steps on-screen visible with the
moment if I scroll to the bottom you’ll see them wanna go ahead and switch out of here
and I’m gonna switch over to a particular topic like a 0 let’s see age-related macular
degeneration there’s a fairly classic slip in here
one of the early ones that was found with the genome-wide association studies that these steps are not necessarily
exclusively related to one particular condition here we see age-related macular
degeneration a couple times over but this one which has the association
also relates to HDL cholesterol and ldl-cholesterol but second graph get this other window it’s a little bit of screen at the
moment by positioning and then using my mouse wheel to scroll
I can zoom in a little bit closer and see some of the a different
associations that are present for this so this particular topic here has all
these red and green slips around it but we’ll see that some other early two
things like longevity and the two cholesterols steps just talked I’m gonna leave age-related macular degeneration instead
search for let’s say cancer will pick bladder cancer versus
thirteen of those and looking at the graph for this one what other things are associated with
the slips because it isn’t as if there’s a step that’s just the bladder cancer said it turns out that we see associations
with colon cancer prostate cancer and liver cancer for so if these whereas other ones only have
a a fewer number associations like gall bladder cancer or gastric its so the grass can be a good way to
get in overview for the association’s ok all the different
medical conditions that are well minute click on Help and then come back
to the mean report so you are too is probably the best way
to really explore your data but sometimes you know that there’s
particular medical condition that you care about or more importantly I think medicines people often look into this genetic
testing and the focus on diseases but genetic diseases are often something
that you kinda just have this increase risk for a reason of a lot you
can do about it your doctor will tell you to you right
exercise don’t smoke but the things that I think are really
valuable are the medicines medicines really
influence decisions that you and your doctor can make together and consider whether or not certain
medicines might go ahead have high risk for you I think for lee
Min does nothing and really jumps out we can see that for
instance for Allegra here she does have one step classified in red and therefore there’s something that she
has she’s a header is i get for it and there’s information here about all
the different drugs this one affects so she’s considering taking Allegra
others resent the click-through really before taking any new medicine I
think it’s worthwhile to consider skimming through here and considering
what kind of reactions you might have its you we can find
something particularly interest well but the most interesting
but a good mix of things %uh club pitted sure I’ll I don’t know
much about it but I can see if there’s most think
classified in this green and good one or two things classified in
grey is not yet classified and red as that and so here’s a few things which you
want to know if she’s about to take that matters and it looks like she’s a C Y P to see
nineteen it’s a particular gene she’s the type 17 ultra-fast metabolize sir and as a
result she breaks down certain medicines more
quickly clicking on more here or the RS numbers as before can tell me more about so what that’ll come back to the main
report show you that it is possible to drill
deeper and deeper and deeper into your data there is a lot of information here but two sections you’ve seen so far
really are probably the the best places to start if you really
want to take a lot deeper there’s also a tab delimited file that
you could open an excel if the reports that we’ve got here are
the right way with Excel or other tools for processing
tab delimited files should really explored anyway you want okay I hope it’s a helpful introduction
to permit yes I hope you’ll consider contributing what
you learn about your genome backed into state PD a help all of us
learn more about ourselves thanks


Add a Comment

Your email address will not be published. Required fields are marked *