Frank Feng wrote:
> Dear all,
>
>
> I don't know much about R, but I now have to run R under Python. From
> a guy who has much experience with R but not know Python at all, I
> get the R statements I need:
>
>> library("hwde") gdata <- data.frame(Population = c("P","P", "P",
>> "N","N","N"), Observed = c(691, 186, 0.5,468, 87, 17 ), locus1 =
>> c("AA", "Aa", "aa", "AA", "Aa", "aa")) hwde(data = gdata)
>
> Run the above in R, then I can get the results I need.
>
> Python is my programming language, so I installed RPy. I have read
> the RPy manual for two days, but I still don't know how to translate
> these R statements to Python statements,
There's your mistake! You don't translate the R statements to Python
statements, you execute the R statements in R...
Suppose you have your data in Python lists called Population,
Observed, and locus1:
>>> from rpy import r
>>> Population = ["P","P", "P", "N","N","N"]
>>> Observed = [691, 186, 0.5,468, 87, 17 ]
>>> locus1 = ["AA", "Aa", "aa", "AA", "Aa", "aa"]
You can transfer data from python variables to R objects with r.assign:
>>> r.assign('Observed',Observed)
[691.0, 186.0, 0.5, 468.0, 87.0, 17.0]
>>> r.assign('Population',Population)
['P', 'P', 'P', 'N', 'N', 'N']
>>> r.assign('locus1',locus1)
['AA', 'Aa', 'aa', 'AA', 'Aa', 'aa']
Now that the data is in R, construct the R data frame object (think of
it like a database table) and call hwde:
>>> library(hwde)
>>>
r('gdata=data.frame(Population=Population,Observed=Observed,locus1=locus1)')
>>> result = r('hwde(gdata)')
[1] "Analysis of Deviance Table"
Resid. Df Resid. Dev Df Deviance
1 5 2155.45
+gp 4 2090.58 1 64.87
+a 3 34.85 1 2055.73
+aa 2 34.75 1 0.10
+gp:a 1 34.75 1 4.442e-03
+gp:aa 0 -6.517e-14 1 34.75
There were 12 warnings (use warnings() to see them)
Now we have a 'result' variable back in python. This is a dict:
>>> result.keys()
['anovatab', 'data.df', 'aovtab.terms']
and somewhere in there will be the numbers you require!
Now, the way I've done this avoids most problems with converting between
python variables and R objects. Python cannot at the moment store R data
frames in python variables and return them safely to R without messing
with the Rpy conversion system that seems to confuse newcomers.
By mucking about with the conversion settings it's much easier - you
can construct the data frame within Python:
>>> import rpy
>>> rpy.set_default_mode(rpy.NO_CONVERSION)
>>>
gdata=r.data_frame(Population=Population,Observed=Observed,locus1=locus1)
>>> result = r.hwde(gdata)
[1] "Analysis of Deviance Table"
Resid. Df Resid. Dev Df Deviance
1 5 2155.45
+gp 4 2090.58 1 64.87
+a 3 34.85 1 2055.73
+aa 2 34.75 1 0.10
+gp:a 1 34.75 1 4.442e-03
+gp:aa 0 -6.517e-14 1 34.75
There were 12 warnings (use warnings() to see them)
but now 'result' is an unconverted R object:
>>> result
<Robj object at 0x00A55280>
But you can convert this back into a dict by switching back to
BASIC_CONVERSION and using the as_py() method:
>>> rpy.set_default_mode(rpy.BASIC_CONVERSION)
>>> result2 = result.as_py()
>>> result2.keys()
['anovatab', 'data.df', 'aovtab.terms']
giving us a Python dict again.
I hope this helps - it certainly helps me to refresh myself on all
this fiddly stuff once every few weeks by answering this kind of query
on the mailing list!
Barry
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
rpy-list mailing list
rpy-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rpy-list
