From: Alec Stephenson <astephen_at_efs.mq.edu.au>
Date: Thu 08 Jul 2004 - 17:25:40 EST

For E:
e <- e[,1] ; uni <- union(c[,1],union(s[,1],h[,1])) Then e[!match(e, uni, 0)] should get you there.

Alec

Alec Stephenson Department of Statistics
Macquarie University
NSW 2109, Australia

>>> S Peri <biocperi@yahoo.com> 07/08/04 02:41pm >>> Dear Alec,
Thank you for your response and it worked. Further, I have another mathematical problem. I apologize ahead as this question is not apt for this list. I am a biologist working at Johns Hopkins School of Medicine.

As I listed in my previous e-mail (attached below)I mentioned I have 4 protein sets. I am now trying to calculate the combinatorics of these sets. My ultimate aim is to draw a venn diagram and find out the proteins that are unique to set C,S,E and H.  I drew a venn diagram graph and I am banging my head to deduce the combinations. It is easy for me to deduce the intersections - that means the protein entries that are present in common. However, it proved very difficult to deduce the following:

I could calculate the following:

```NP_*** present in both C and E (C intersection E)
NP_*** present in both C and H (C intersection H)
NP_*** present in both C and S (C ^ S)
NP_*** present in both E and H (E ^ H)
NP_*** present in both E and S (E ^ S)
NP_*** present in both H and S (H ^ S)

NP_*** present in C, E and H (C^E^H)
NP_*** present in C, H and S (C^H^S)
```

NP_*** present in E, H and S (E^H^S)
NP_*** present in E, S and C (E^S^C)

It is very difficult for me to deduce the following:

```NP_**** entries specific to E
NP_**** entries specific to H
NP_**** entries specific to S

```

I waster many pages but could not derive some solution to get unique elements for sets, s,e,h,and C.

Can any one help me by suggesting some way to get these.

Thank you and I apologise again for posting the wrong question.

SP

• S Peri <biocperi@yahoo.com> wrote:
> Hello group,
> I am learning R and I am new to many concepts.I
> face
> the following errors when I am trying to execute the
> following. I have 4 text files with protein
> accession
> numbers. I wanted to represent them in a venn
> diagram
> and for that I using intersect and setdiff
> functions.
>
> My data looks like this:
>
> file1.txt (c):
> NP_000005
> NP_000020
> NP_000030
> NP_000053
>
> file2.txt(e):
> NP_000005
> NP_000020
> NP_000030
> NP_000031
> NP_000053
> NP_000055
> NP_000087
>
> file3.txt(h):
> NP_000005
> NP_000020
> NP_000030
> NP_000053
> NP_000055
> NP_000057
> NP_000087
>
> file4.txt (s):
> NP_000005
> NP_000020
> NP_000030
> NP_000033
> NP_000053
> NP_000055
> NP_000087
> NP_000168
>
>
> Now I did the following FIRST time:
>
> > class(c)
> [1] "data.frame"
> > class(s)
> [1] "data.frame"
> > CiS=intersect(c,s)
> > CiS
> NULL data frame with 0 rows
> ##### Why am I getting NULL data error. I know there
> are common elements between c and S. ##########
>
> Error in unique(y[match(x, y, 0)]) : Argument "y" is
> missing, with no default
> Error in unique(y[match(x, y, 0)]) : Argument "y" is
> missing, with no default
>
> ##### Why am I getting this error.
>
>
>
> Second thing I did:
>
> Again I never get intersection of C,E and S,H.
>
>
> thank you
>
> SP
>
> ______________________________________________
> R-help@stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help