reduce space between X-axis ticks

Question

[rohitsatyam102]

Hi I am trying to make a bubble plot for enriched pathway in R using the code given below but I wish to reduce the space between the x-axis ticks. When I use facet_grid with space = "free_y" the theme(aspect.ratio = 0.4) gives me the following error:

Error in `f()`:
! Free scales cannot be mixed with a fixed aspect ratio
Run `rlang::last_error()` to see where the error occurred.

From discussion posted here and here I understand that it is difficult to achieve but is there a way that I can edit the ggplot2 object to reduce the space between x -axis ticks

ggplot(df, aes(yend=Term,y = Term, xend = Category, x = Category,colour = -log10(Adjusted.P.value), fill= -log10(Adjusted.P.value), size=count))+
 geom_point()+
scale_color_viridis_c(option = "C") + 
scale_fill_viridis_c(option="C") +
ggthemes::theme_few() + 
scale_y_discrete(labels=function(x) stringr::str_wrap(x,width = 55))+
labs(x=NULL, y = NULL)+
theme(aspect.ratio = 0.4)+
facet_grid(firstannotation ~ ., scales = "free",switch = "y", space = "free")+
theme(axis.text.x=element_text(color = "black", size=10, face="bold", angle=90, hjust = 1),axis.text.y=element_text(color = "black", size=9, face="bold"),
strip.placement = "outside",strip.text.y.left = element_text(angle = 0),strip.background.y=element_rect(color = NA,  fill="grey"))

Data look like:

Term	Category	Overlap	P.value	Adjusted.P.value	Old.P.value	Combined.Score	Genes	Gene.Ratio	label	firstannotation
Protein metabolism	Exp_20h	54/442	1.30E-07	2.38E-05	2.348932	37.24151	EIF4A1;TRAM1;RPL32;RPL31;TUSC3;GALNT15;GALNT18;RPL6;RPS4X;STS;RPL18A;RPL36;RPL35;RPLP2;RPL38;RPL37;EIF5B;ALG9;RPL21;RPS8;RPL23;RPL22;RPS3A;ALG11;B3GNT5;RPL37A;RPL24;RPL27;RPL26;GAS6;RPL28;PIGF;PFDN5;RPL12;RPL11;RPS4Y1;MAN2A1;GRPEL2;RPL13;GALNT7;SEC24B;GALNT1;C1GALT1;RPL35A;RPL23A;TBCA;RPS26;RPS25;RPS27;EIF3K;RPL27A;TOMM7;RPS20;RPS24	6	Metabolism,Protein folding, sorting and degradation,Protein modification,Translation	Metabolism
Cytoplasmic ribosomal proteins	Exp_20h	34/108	8.75E-17	9.62E-14	7.74482	286.36	RPL32;RPL31;RPL12;RPL11;RPS4Y1;RPL6;RPS4X;RPS6KA3;RPL18A;RPL36;RPL35;RPLP2;RPL13;RPL38;RPL37;RPL21;RPS8;RPL23;RPL22;RPL35A;RPS3A;RPL23A;RPS26;RPS25;RPS27;EIF3K;RPL27A;RPL37A;RPL24;RPL27;RPS20;RPL26;RPL28;RPS24	1	Genetic Information Processing:Translation,Genetic disease,Nucleic acid metabolism,Translation,Transport	Genetic Information Processing
Translation	Exp_20h	38/151	5.24E-15	1.92E-12	5.677127	186.6828	EIF4A1;TRAM1;RPL32;RPL31;RPL12;RPL11;RPS27L;RPS4Y1;RPL6;RPS4X;RPL18A;RPL36;RPL35;RPLP2;RPL13;RPL38;RPL37;EIF5B;RPL21;RPS8;RPL23;RPL22;RPL35A;RPS3A;RPL23A;RPS26;RPS25;RPS27;EIF3K;RPL27A;RPL37A;RPL24;RPL27;RPS20;RPL22L1;RPL26;RPL28;RPS24	3	Genetic Information Processing:Translation,Translation	Genetic Information Processing
TGF-beta regulation of extracellular matrix	Exp_2h	34/565	1.37E-10	1.67E-08	4.056584	92.13191	CXCL6;SYNM;CSF1;C1S;ELN;SLC40A1;CXCL1;PTGS2;FBLN5;ADAMTS1;CTSK;HAS2;CD36;JUNB;MAP3K5;KLF10;POSTN;VCAM1;PTGIS;PLCL1;PLA2G4C;BTBD11;INHBA;DKK1;ARL4A;COL1A2;ID2;STK17B;DDIT4;FGF18;CYP1A1;TRIB3;PRKD1;STEAP2	6	Genetic disease,Signal transduction	Genetic disease
TGF-beta regulation of extracellular matrix	Exp_2h	54/565	0.000168	0.020536	1.771653	15.3966	RB1;CNTNAP1;ELN;GOLIM4;SPINT2;PTPRK;SLC7A11;RPL18A;CTSK;LEPR;NRCAM;CD36;NCK1;KLF10;POSTN;IL1R1;PLA2G4C;NSMCE1;RBL2;RBP1;ANG;PDE5A;IDI1;CUL5;CEBPD;PIK3R1;PTGS2;PPL;NID2;SELENBP1;SDCBP;PDGFD;APOE;S100A10;SVIL;CDKN2B;JAG1;PLCL1;PCDHGC3;AKR1C3;PDP1;BMP4;TF;COL1A2;TSPAN13;NFIB;TNFSF4;ID2;QPCT;RAB13;CYP1A1;PIR;ACKR4;EVI2B	9	Genetic disease,Signal transduction	Genetic disease
TGF-beta regulation of extracellular matrix	Exp_2h	31/565	8.64E-13	1.08E-10	5.499825	152.7713	CXCL6;SYNM;CSF1;SLC40A1;STC1;TNFAIP3;CXCL1;PTGS2;CYR61;MYC;CTSK;CD36;JUNB;MAP3K5;KLF10;EDN1;POSTN;VCAM1;PLA2G4C;BTBD11;APLN;BMP4;ARL4A;COL1A2;STK17B;CYP1A1;ACKR3;TRIB3;PRKD1;STEAP2;ABCG1	5	Genetic disease,Signal transduction	Genetic disease

[Matthias Zepper]

Can you please show a head() of the data you are trying to plot?

This is really a wild assortment of mappings

aes(yend=Term,y = Term, xend = Category, x = Category,colour = -log10(Adjusted.P.value), fill= -log10(Adjusted.P.value), size=count))

and I think with a bit of preprocessing of your input data, you can simplify a lot in the actual plot.

[rohitsatyam102]

Hi Matthias Zepper I have added the data above for your reference.

[Matthias Zepper]

library("ggplot2")

df <- data.frame(metacategory=sample(c("A","B"),20,replace=TRUE),
                 category_a=rep(LETTERS[1:5],4),
                 category_b=rep(LETTERS[1:5],each=4),
                 value=rnbinom(20,0.4,0.2))

ggplot(df,aes(x=category_a,y=category_b,size=value)) + geom_point() + facet_grid(metacategory ~ . )

is a minimal example for the type of plot you are trying to create. Make sure, that your df has three discrete columns (as factors) that you would like to map to x,y and for the grid. Map a numeric value to size and if you like, you can also map the p-value to color or maybe to alpha (1 - p.value then of course, such that the significant ones are shown).

[rohitsatyam102]

I understand that and maybe if you give a closer look to my code I used somtething similar. The problem is when I do

ggplot(df,aes(x=category_a,y=category_b,size=value)) + geom_point() + facet_grid(metacategory ~ . , scales = "free",switch = "y", space = "free")+coord_fixed()

or

ggplot(df,aes(x=category_a,y=category_b,size=value)) + geom_point() + facet_grid(metacategory ~ . , scales = "free",switch = "y", space = "free")+theme(aspect.ratio = 0.4)

I get the following errors:

error in `f()`:
! Free scales cannot be mixed with a fixed aspect ratio

or

Error in `f()`:
! coord_fixed doesn't support free scales
Run `rlang::last_error()` to see where the error occurred.

I want something like this where the x-axis ticks are equidistant and closer to each other

Answer 1

[Kevin Blighe]

Just alter the dimensions of the output device. If you are exporting to PDF, for example, alter the values of width and height.

Kevin

[rohitsatyam102]

Yes eventually I took that path. I thought if there was a more controlled way of do it before saving the file rather than altering the width and height.