facets work in qplot but facet_wrap produces an error in ggplot

Hello all,

I am having a perplexing problem trying to use facet_wrap in ggplot,
with
both my real dataset and a simplified dummy dataset.  I am trying to
plot
heterozygosity across the genome for multiple individuals, with each
chromosome shown separately.

My dummy data:
chr1	123000	124000	2	0.00002	26	0.00026	indiv1
chr1	124000	125000	3	0.00003	12	0.00012	indiv1
chr1	125000	126000	1	0.00001	6	0.00006	indiv1
chr1	126000	126000	2	0.00002	14	0.00014	indiv1
chr2	123000	124000	6	0.00006	20	0.00020	indiv1
chr2	124000	125000	0	0.00000	12	0.00012	indiv1
chr1	123000	124000	2	0.00002	26	0.00026	indiv2
chr1	124000	125000	3	0.00003	12	0.00012	indiv2
chr1	125000	126000	1	0.00001	6	0.00006	indiv2
chr1	126000	126000	2	0.00002	14	0.00014	indiv2
chr2	123000	124000	6	0.00006	20	0.00020	indiv2
chr2	124000	125000	0	0.00000	12	0.00012	indiv2

My code to read in the data:
hetshoms <- read.table("fakedata.txt", header=F)

chrom <- hetshoms$V1
start.pos <- hetshoms$V2
end.pos <- hetshoms$V3
hets <- hetshoms$V4
het_stat <- hetshoms$V5
homs <- hetshoms$V6
hom_stat <- hetshoms$V7
indiv <- hetshoms$V8

HetRatio <- hets/(hets+homs)

When I try to plot the chromosomes separately in qplot, it works fine:
testplot <- qplot(start.pos, HetRatio, facets = chrom ~ .,
colour=chrom)

But when I try an analogous thing in ggplot, it does not work.
The first part works fine:
testplot <- ggplot(hetshoms, aes(x=start.pos, y=HetRatio)) +
geom_point(aes(color=chrom, alpha=1/4)) 

but when I try to add the facet_wrap:
testplot + facet_wrap(~chrom)

This produces the following error (and no plot)
"Error en layout_base(data, vars, drop = drop) : 
 At least one layer must contain all variables used for facetting"

I have tried adding an (as.formula(paste)) and directly calling
hetshoms$V1
but neither solves the problem.

Can anyone please point out where I have gone wrong and how to fix my
code?  

Much appreciated,
Loren






--
View this message in context:
http://r.789695.n4.nabble.com/facets-work-in-qplot-but-facet-wrap-produces-an-error-in-ggplot-tp4705058.html
Sent from the R help mailing list archive at Nabble.com.

You MUST put all data you plan to refer to into a data frame when using
ggplot. There are a couple of ways you could do this... the easiest is to
put a header line in the data file with column names. Or, you can assign a
vector of new names to the names of the data frame.

names( hetshoms ) <- c( "chrom", "start.pos", "end.pos", "hets",
"het_stat", "homs", "hom_stat", "indiv" )
---------------------------------------------------------------------------
Jeff Newmiller                        The     .....       .....  Go Live...
DCN:<jdnewmil at dcn.davis.ca.us>        Basics: ##.#.       ##.#.  Live
Go...
                                      Live:   OO#.. Dead: OO#..  Playing
Research Engineer (Solar/Batteries            O.O#.       #.O#.  with
/Software/Embedded Controllers)               .OO#.       .OO#.  rocks...1k
---------------------------------------------------------------------------
Sent from my phone. Please excuse my brevity.

On March 24, 2015 3:40:11 PM PDT, Loren <sackettl at colorado.edu> wrote:

Hello all,

I am having a perplexing problem trying to use facet_wrap in ggplot,
with
both my real dataset and a simplified dummy dataset.  I am trying to
plot
heterozygosity across the genome for multiple individuals, with each
chromosome shown separately.

My dummy data:
chr1   123000  124000  2       0.00002 26      0.00026 indiv1
chr1   124000  125000  3       0.00003 12      0.00012 indiv1
chr1   125000  126000  1       0.00001 6       0.00006 indiv1
chr1   126000  126000  2       0.00002 14      0.00014 indiv1
chr2   123000  124000  6       0.00006 20      0.00020 indiv1
chr2   124000  125000  0       0.00000 12      0.00012 indiv1
chr1   123000  124000  2       0.00002 26      0.00026 indiv2
chr1   124000  125000  3       0.00003 12      0.00012 indiv2
chr1   125000  126000  1       0.00001 6       0.00006 indiv2
chr1   126000  126000  2       0.00002 14      0.00014 indiv2
chr2   123000  124000  6       0.00006 20      0.00020 indiv2
chr2   124000  125000  0       0.00000 12      0.00012 indiv2

My code to read in the data:
hetshoms <- read.table("fakedata.txt", header=F)

chrom <- hetshoms$V1
start.pos <- hetshoms$V2
end.pos <- hetshoms$V3
hets <- hetshoms$V4
het_stat <- hetshoms$V5
homs <- hetshoms$V6
hom_stat <- hetshoms$V7
indiv <- hetshoms$V8

HetRatio <- hets/(hets+homs)

When I try to plot the chromosomes separately in qplot, it works fine:
testplot <- qplot(start.pos, HetRatio, facets = chrom ~ .,
colour=chrom)

But when I try an analogous thing in ggplot, it does not work.
The first part works fine:
testplot <- ggplot(hetshoms, aes(x=start.pos, y=HetRatio)) +
geom_point(aes(color=chrom, alpha=1/4))

but when I try to add the facet_wrap:
testplot + facet_wrap(~chrom)

This produces the following error (and no plot)
"Error en layout_base(data, vars, drop = drop) :
 At least one layer must contain all variables used for facetting"

I have tried adding an (as.formula(paste)) and directly calling
hetshoms$V1
but neither solves the problem.

Can anyone please point out where I have gone wrong and how to fix my
code?

Much appreciated,
Loren






--
View this message in context:

http://r.789695.n4.nabble.com/facets-work-in-qplot-but-facet-wrap-produces-an-error-in-ggplot-tp4705058.html

Sent from the R help mailing list archive at Nabble.com.

Thanks, Jeff.  I tried this previously by using a header in my data
file
(and 'header=TRUE'), but for some reason, that did not seem to work.

Creating a 'names' vector as you suggested did solve the problem,
though.

Thank you!
Loren


2015-03-24 23:19 GMT-04:00 Jeff Newmiller <jdnewmil at dcn.davis.ca.us>:

You MUST put all data you plan to refer to into a data frame when

using

ggplot. There are a couple of ways you could do this... the easiest

is to

put a header line in the data file with column names. Or, you can

assign a

vector of new names to the names of the data frame.

names( hetshoms ) <- c( "chrom", "start.pos", "end.pos", "hets",
"het_stat", "homs", "hom_stat", "indiv" )

---------------------------------------------------------------------------

Jeff Newmiller                        The     .....       .....  Go

Live...

DCN:<jdnewmil at dcn.davis.ca.us>        Basics: ##.#.       ##.#.  Live
Go...
                                      Live:   OO#.. Dead: OO#.. 

Playing

Research Engineer (Solar/Batteries            O.O#.       #.O#.  with
/Software/Embedded Controllers)               .OO#.       .OO#. 

rocks...1k

---------------------------------------------------------------------------

Sent from my phone. Please excuse my brevity.

On March 24, 2015 3:40:11 PM PDT, Loren <sackettl at colorado.edu>

wrote:

Hello all,

I am having a perplexing problem trying to use facet_wrap in ggplot,
with
both my real dataset and a simplified dummy dataset.  I am trying to
plot
heterozygosity across the genome for multiple individuals, with each
chromosome shown separately.

My dummy data:
chr1   123000  124000  2       0.00002 26      0.00026 indiv1
chr1   124000  125000  3       0.00003 12      0.00012 indiv1
chr1   125000  126000  1       0.00001 6       0.00006 indiv1
chr1   126000  126000  2       0.00002 14      0.00014 indiv1
chr2   123000  124000  6       0.00006 20      0.00020 indiv1
chr2   124000  125000  0       0.00000 12      0.00012 indiv1
chr1   123000  124000  2       0.00002 26      0.00026 indiv2
chr1   124000  125000  3       0.00003 12      0.00012 indiv2
chr1   125000  126000  1       0.00001 6       0.00006 indiv2
chr1   126000  126000  2       0.00002 14      0.00014 indiv2
chr2   123000  124000  6       0.00006 20      0.00020 indiv2
chr2   124000  125000  0       0.00000 12      0.00012 indiv2

My code to read in the data:
hetshoms <- read.table("fakedata.txt", header=F)

chrom <- hetshoms$V1
start.pos <- hetshoms$V2
end.pos <- hetshoms$V3
hets <- hetshoms$V4
het_stat <- hetshoms$V5
homs <- hetshoms$V6
hom_stat <- hetshoms$V7
indiv <- hetshoms$V8

HetRatio <- hets/(hets+homs)

When I try to plot the chromosomes separately in qplot, it works

fine:

testplot <- qplot(start.pos, HetRatio, facets = chrom ~ .,
colour=chrom)

But when I try an analogous thing in ggplot, it does not work.
The first part works fine:
testplot <- ggplot(hetshoms, aes(x=start.pos, y=HetRatio)) +
geom_point(aes(color=chrom, alpha=1/4))

but when I try to add the facet_wrap:
testplot + facet_wrap(~chrom)

This produces the following error (and no plot)
"Error en layout_base(data, vars, drop = drop) :
 At least one layer must contain all variables used for facetting"

I have tried adding an (as.formula(paste)) and directly calling
hetshoms$V1
but neither solves the problem.

Can anyone please point out where I have gone wrong and how to fix

my

code?

Much appreciated,
Loren






--
View this message in context:

http://r.789695.n4.nabble.com/facets-work-in-qplot-but-facet-wrap-produces-an-error-in-ggplot-tp4705058.html

Sent from the R help mailing list archive at Nabble.com.