[R] rbinom and probability

[David Winsemius]

On Feb 4, 2011, at 4:24 PM, Mcmahon, Kwyatt wrote:

> Hello compadRes,
>
> I'm developing a script that selects "cells" over a certain  
> metabolic rate to kill them.  A rate between 9 and 12 means that the  
> cells are candidates for death.
>
> I'll show you what I mean:
>
> # a would be a vector of cell metabolic rates.
> a<-c(8, 7, 9, 6, 10, 11, 4, 5, 6)
>
> #now identify which cells will be candidates for death, namely those  
> cells with metabolic rates>9
> b<-a[a>6]
> #This would select all cells with rates >6
> rbinom(length(b), size=1, prob=1)

No, you have an overly complex way of generating a vector of all 1's  
with length(b). And it's not clear what you mean by "select". I see no  
caode that would do any selection or indexing.

>
> #This will select cells using a an exponential probability  
> distribution
> rbinom(length(b), size=1, prob=(seq(0, 1, by=0.1)^2))

Again no selection appears to be taking place.

>
> I have two questions:
>
> 1)      Am I correct in my interpretations?
>
> 2)      I'd actually like to have this probability scaled so that  
> cells with rates of 9 are unlikely to be selected and those near 12  
> are highly likely.  How can I code this?

a[ sample(1:length(a), prob=0.1+(a>9) ,replace=TRUE) ]
[1] 10 10 11 10 10 11 10 11 10
 > a[sample(1:length(a), prob=0.1+(a>9) ,replace=TRUE) ]
[1] 10 10 11 11 11 11  5 10 10

The "rates" over 9 are 11 times  (= 1.1/0.1) as likely to be selected.  
You can adjust the ratio by altering the 0.1 value."+" with the first  
argument numeric act to coerce the logical vector to 1's and 0's.

-- 
David.


>
> Thanks in advance,
>
> Wyatt
>
>> sessionInfo()
> R version 2.11.1 (2010-05-31)
> i386-pc-mingw32
>
> locale:
> [1] LC_COLLATE=English_United States.1252
> [2] LC_CTYPE=English_United States.1252
> [3] LC_MONETARY=English_United States.1252
> [4] LC_NUMERIC=C
> [5] LC_TIME=English_United States.1252
>
> attached base packages:
> [1] stats     graphics  grDevices utils     datasets  methods   base
>
> loaded via a namespace (and not attached):
> [1] tools_2.11.1
>
>
>
> K. Wyatt McMahon, Ph.D.
> Postdoctoral Research Associate
> Department of Internal Medicine
> 3601 4th St. - Lubbock, TX - 79430
> P: 806-743-4072
> F: 806-743-3148
>
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

David Winsemius, MD
West Hartford, CT