Extraction of total RNA from single-oocytes and single-cell mRNA sequencing of swine oocytes
Katelyn M. Kimble, Sarah E. Dickinson, Fernando H. Biase
January/2018
Overview
This code accompanies the paper Kimble et al. (2018) Extraction of total RNA from single-oocytes and full-length single-cell mRNA sequencing of swine oocytes, as Supplementary file 1. Code created by Fernando Biase. Please, direct questions to fbiase at auburn dot edu.
An updated version of this code may be available on www.biaselaboratory.com .
Code
Load the necessary libraries.
library(knitr)
library(Rsamtools)
library(stringr)
library(gplots)
library(ggplot2)
library(ggpubr)
library(scales)
library(kableExtra)
library(biomaRt)
library(RColorBrewer)#function for collapsing the list of lists into a single list
#as per the Rsamtools vignette
.unlist <- function (x){
## do.call(c, ...) coerces factor to integer, which is undesired
x1 <- x[[1L]]
if (is.factor(x1)){
structure(unlist(x), class = "factor", levels = levels(x1))
} else {
do.call(c, x)
}
}Load the annotations from Ensembl.
mart = useMart("ensembl", dataset = "sscrofa_gene_ensembl",host="www.ensembl.org")
annotation.genelength.biomart<-getBM(attributes = c("ensembl_gene_id","ensembl_transcript_id", "transcript_length","gene_biotype"), mart = mart)
annotation.genelength.biomart<-annotation.genelength.biomart[annotation.genelength.biomart$gene_biotype =="protein_coding",]
annotation.genelength.biomart<-annotation.genelength.biomart[order(annotation.genelength.biomart$ensembl_gene_id, -annotation.genelength.biomart$transcript_length),]
#head(annotation.genelength.biomart[duplicated(annotation.genelength.biomart$ensembl_gene_id),])
annotation.genelength.biomart<-annotation.genelength.biomart[annotation.genelength.biomart$transcript_length > 400,]
annotation.genelength.biomart.a<-annotation.genelength.biomart[!duplicated(annotation.genelength.biomart$ensembl_gene_id),]annotation.gene.biomart<-getBM(attributes = c("ensembl_gene_id", "hgnc_symbol","external_gene_name", "description","gene_biotype"), mart = mart)
annotation.gene.biomart<-annotation.gene.biomart[annotation.gene.biomart$gene_biotype =="protein_coding",]Load bam files and generate the objects with data.
data_fpkm <- list()
data_coverage <- list()
bamInPath<-"/home/fernando/oocyte_cumulus_pig/alignment_oocyte_cdna"
#read in BAM file
i<-1
for (sample in c('SL269995','SL269992','SL270021','SL270003','SL269998')){
bam <- scanBam(paste(bamInPath,sample,"Aligned.out.merged.sorted.bam",sep="/"))
#store names of BAM fields
bam_field <- names(bam[[1]])
#go through each BAM field and unlist
list <- lapply(bam_field, function(y) .unlist(lapply(bam, "[[", y)))
#store as data frame
bam_df <- do.call("DataFrame", list)
names(bam_df) <- bam_field
bam_df<-as.data.frame(bam_df[,c(3,5,10)])
bam_df$transcript<-str_split_fixed(bam_df$rname, "\\.",2)[,1]
bam_df<-bam_df[,c(4,2,3)]
total_fragments_aligned<-dim(bam_df)[1]
bam_df_not_duplicated<-bam_df[bam_df$transcript %in% annotation.genelength.biomart.a$ensembl_transcript_id,]
table_transcript_count<-data.frame(table(bam_df_not_duplicated$transcript))
table_transcript_count<-merge(table_transcript_count, annotation.genelength.biomart.a, by.x="Var1", by.y="ensembl_transcript_id", all=FALSE)
table_transcript_count$fpkm<-( ( (table_transcript_count$Freq) / ((table_transcript_count$transcript_length/1000) * (total_fragments_aligned/1000000)) ) )
table_transcript_count<-table_transcript_count[,c(3,6)]
colnames(table_transcript_count)<-c("ensembl_gene_id", paste(sample))
data_fpkm[[i]]<-table_transcript_count
bam_df1<-bam_df[,c(1,2)]
bam_df2<-bam_df[,c(1,3)]
colnames(bam_df1)<-c("transcript","position")
colnames(bam_df2)<-c("transcript","position")
bam_df3<-rbind(bam_df1,bam_df2)
bam_df3<-bam_df3[!is.na(bam_df3$position),]
data_coverage[[i]]<-bam_df3
i<- i+1
rm(bam,bam_df,bam_df1,bam_df2,bam_df3,bam_df_not_duplicated, list)
}Create the object that contains the expression data (FPKM) for the genes detected in all oocytes. The code used for saving the file is commented (#) and the table is printed below.
data_fpkm_a<-Reduce(function(x, y) merge(x, y, all=TRUE), data_fpkm)
rownames(data_fpkm_a)<-data_fpkm_a$ensembl_gene_id
data_fpkm_a<-data_fpkm_a[, c(2:dim(data_fpkm_a)[2])]
data_fpkm_b<-data_fpkm_a[rowSums(data_fpkm_a > 0.3) >=5,]
data_fpkm_b<-data_fpkm_b[complete.cases(data_fpkm_b),]
data_fpkm_b_annotated<-merge(data_fpkm_b, annotation.gene.biomart, by.x= "row.names", by.y="ensembl_gene_id", all.x=TRUE, all.y=FALSE)
colnames(data_fpkm_b_annotated)<-c("Ensembl ID","oocyte 1","oocyte 2","oocyte 3","oocyte 4","oocyte 5","hgnc_symbol","external_gene_name","description","gene_biotype")Supplementary file 2.
kable(head(data_fpkm_b_annotated), format = "html") %>% kable_styling(bootstrap_options = c("striped", "hover", "condensed"),full_width =TRUE,font_size = 10)| Ensembl ID | oocyte 1 | oocyte 2 | oocyte 3 | oocyte 4 | oocyte 5 | hgnc_symbol | external_gene_name | description | gene_biotype |
|---|---|---|---|---|---|---|---|---|---|
| ENSSSCG00000000002 | 3.054407 | 1.673888 | 1.5659515 | 1.370527 | 1.782784 | GTSE1 | GTSE1 | G2 and S-phase expressed 1 [Source:HGNC Symbol;Acc:HGNC:13698] | protein_coding |
| ENSSSCG00000000005 | 3.272916 | 8.486265 | 3.1585472 | 4.837654 | 2.376952 | CDPF1 | CDPF1 | cysteine rich DPF motif domain containing 1 [Source:HGNC Symbol;Acc:HGNC:33710] | protein_coding |
| ENSSSCG00000000006 | 17.612150 | 17.479475 | 18.5034100 | 4.667341 | 18.522566 | PPARA | Sus scrofa peroxisome proliferator activated receptor alpha (PPARA), mRNA. [Source:RefSeq mRNA;Acc:NM_001044526] | protein_coding | |
| ENSSSCG00000000007 | 8.155091 | 11.088408 | 10.7312353 | 12.258417 | 8.616283 | TRMU | TRMU | tRNA 5-methylaminomethyl-2-thiouridylate methyltransferase [Source:HGNC Symbol;Acc:HGNC:25481] | protein_coding |
| ENSSSCG00000000010 | 2.036271 | 1.823013 | 3.5314462 | 6.023551 | 1.844337 | FBLN1 | FBLN1 | fibulin 1 [Source:HGNC Symbol;Acc:HGNC:3600] | protein_coding |
| ENSSSCG00000000014 | 1.089129 | 0.435553 | 0.7445085 | 1.954791 | 1.228299 | FAM118A | FAM118A | family with sequence similarity 118 member A [Source:HGNC Symbol;Acc:HGNC:1313] | protein_coding |
data_fpkm_a[is.na(data_fpkm_a)]<-0
heatmap.2(log10(as.matrix(data_fpkm_a+1)),
labRow=FALSE,
labCol=FALSE,
scale='none',
trace='none',
keysize=1,
key.title='Log10(FPKM+1)',
col=colorRampPalette(c("white", "blue4"))(n = 256),
margins=c(0,0),
key.par = list(cex=0.55))
data_coverage_a<-do.call("rbind", data_coverage)
data_coverage_a$sample<-c(rep(c("oocyte_1","oocyte_2","oocyte_3","oocyte_4","oocyte_5"),
c(dim(data_coverage[[1]])[1], dim(data_coverage[[2]])[1],dim(data_coverage[[3]])[1],dim(data_coverage[[4]])[1],dim(data_coverage[[5]])[1] )) )
data_coverage_a<-merge(data_coverage_a, annotation.genelength.biomart.a, by.x="transcript", by.y="ensembl_transcript_id", all=FALSE)
data_coverage_a<-data_coverage_a[data_coverage_a$ensembl_gene_id %in% rownames(data_fpkm_b),]
data_coverage_a$percentage<-data_coverage_a$position/data_coverage_a$transcript_length * 100
data_coverage_a<-data_coverage_a[data_coverage_a$percentage >0 & data_coverage_a$percentage < 100 ,]Code to generate the individual images observed on Figure 2 on the paper.
font_1<-8
font_2<-6
data_fpkm_c<-merge(data_fpkm_b, annotation.genelength.biomart.a, by.x='row.names', by.y= 'ensembl_gene_id', all.x=TRUE, all.y=FALSE)
data_fpkm_c<-data_fpkm_c[,c(7,8)]
data_fpkm_c$sample<-"oocyte"
data_fpkm_c<-rbind(data_fpkm_c, data.frame('ensembl_transcript_id'=annotation.genelength.biomart.a$ensembl_gene_id, 'transcript_length'=annotation.genelength.biomart.a$transcript_length, sample='ensembl'))
plot1<-ggplot(data_fpkm_c, aes(x = sample, y = transcript_length)) +
geom_boxplot() +
scale_y_continuous(name = "transcript length (nt)")+
scale_x_discrete(name = NULL, labels=c('Ensembl', 'porcine\noocyte')) +
theme_bw()+
theme(
panel.background = element_blank(),
plot.background = element_blank(),
panel.grid =element_blank(),
axis.text =element_text(colour="black", size = font_1),
axis.title.y =element_text(colour="black", size = font_1, vjust=0, hjust=0.5)
)
plot2<-ggplot(data_coverage_a, aes(x = percentage)) +
geom_density()+
theme_bw()+
theme(
panel.background = element_blank(),
plot.background = element_blank(),
panel.grid =element_blank(),
axis.text =element_text(colour="black", size = font_1),
axis.title =element_text(colour="black", size = font_1)
)
data_coverage_b<-data_coverage_a[data_coverage_a$transcript_length >10000 & data_coverage_a$transcript_length<15000,]
data_coverage_c<-data.frame(
'x_axis' = c( rep(density(data_coverage_b[data_coverage_b$sample=="oocyte_1",7])$x , 5)),
'density'= c( density(data_coverage_b[data_coverage_b$sample=="oocyte_1",7])$y, density(data_coverage_b[data_coverage_b$sample=="oocyte_2",7])$y,
density(data_coverage_b[data_coverage_b$sample=="oocyte_3",7])$y, density(data_coverage_b[data_coverage_b$sample=="oocyte_4",7])$y,
density(data_coverage_b[data_coverage_b$sample=="oocyte_5",7])$y),
'sample' = c(rep(c("oocyte_1","oocyte_2","oocyte_3","oocyte_4","oocyte_5"),
c(
length(density(data_coverage_b[data_coverage_b$sample=="oocyte_1",7])$y), length(density(data_coverage_b[data_coverage_b$sample=="oocyte_2",7])$y),
length(density(data_coverage_b[data_coverage_b$sample=="oocyte_3",7])$y), length(density(data_coverage_b[data_coverage_b$sample=="oocyte_4",7])$y),
length(density(data_coverage_b[data_coverage_b$sample=="oocyte_5",7])$y))
)
))
plot3<-ggplot(data_coverage_c, aes(x_axis, as.factor(sample))) +
geom_tile(aes(fill = density),colour = "white")+
scale_fill_gradient(low = "#F0F0F0",high = "#000000")+
ggtitle("transcript length 10,000 - 15,000 nt (80 genes)") +
theme(
panel.background = element_blank(),
plot.background = element_blank(),
panel.grid =element_blank(),
plot.title =element_text(size=font_1, color="black",vjust = 0.4,hjust = 0.5, margin=margin(t =0, r = 0, b = 0, l = 0, unit = "pt")),
axis.text.y =element_text(size=font_1, color="black"),
axis.text.x =element_blank(),
axis.title =element_blank(),
axis.ticks =element_blank(),
axis.line =element_blank(),
legend.title =element_text(size=font_2),
legend.text =element_text(size=font_2),
plot.margin =margin(t = 0, r = 0, b = 0, l = 0, unit = "pt"),
legend.key.height=unit(0.2, "cm"),
legend.key.width=unit(0.2, "cm")
)
data_coverage_b<-data_coverage_a[data_coverage_a$transcript_length >5000 & data_coverage_a$transcript_length<10000,]
data_coverage_c<-data.frame(
'x_axis' = c( rep(density(data_coverage_b[data_coverage_b$sample=="oocyte_1",7])$x , 5)),
'density'= c( density(data_coverage_b[data_coverage_b$sample=="oocyte_1",7])$y, density(data_coverage_b[data_coverage_b$sample=="oocyte_2",7])$y,
density(data_coverage_b[data_coverage_b$sample=="oocyte_3",7])$y, density(data_coverage_b[data_coverage_b$sample=="oocyte_4",7])$y,
density(data_coverage_b[data_coverage_b$sample=="oocyte_5",7])$y),
'sample' = c(rep(c("oocyte_1","oocyte_2","oocyte_3","oocyte_4","oocyte_5"),
c(
length(density(data_coverage_b[data_coverage_b$sample=="oocyte_1",7])$y), length(density(data_coverage_b[data_coverage_b$sample=="oocyte_2",7])$y),
length(density(data_coverage_b[data_coverage_b$sample=="oocyte_3",7])$y), length(density(data_coverage_b[data_coverage_b$sample=="oocyte_4",7])$y),
length(density(data_coverage_b[data_coverage_b$sample=="oocyte_5",7])$y))
)
))
plot4<-ggplot(data_coverage_c, aes(x_axis, as.factor(sample))) +
geom_tile(aes(fill = density),colour = "white")+
scale_fill_gradient(low = "#F0F0F0",high = "#000000")+
ggtitle("transcript length 5,000 - 10,000 nt (2,888 genes)") +
theme(
panel.background = element_blank(),
plot.background = element_blank(),
panel.grid =element_blank(),
plot.title =element_text(size=font_1, color="black",vjust = 0.4,hjust = 0.5, margin=margin(t =0, r = 0, b = 0, l = 0, unit = "pt")),
axis.text.y =element_text(size=font_1, color="black"),
axis.text.x =element_blank(),
axis.title =element_blank(),
axis.ticks =element_blank(),
axis.line =element_blank(),
legend.title =element_text(size=font_2),
legend.text =element_text(size=font_2),
plot.margin =margin(t = 0, r = 0, b = 0, l = 0, unit = "pt"),
legend.key.height=unit(0.2, "cm"),
legend.key.width=unit(0.2, "cm")
)
data_coverage_b<-data_coverage_a[ data_coverage_a$transcript_length<5000,]
data_coverage_c<-data.frame(
'x_axis' = c( rep(density(data_coverage_b[data_coverage_b$sample=="oocyte_1",7])$x , 5)),
'density'= c( density(data_coverage_b[data_coverage_b$sample=="oocyte_1",7])$y, density(data_coverage_b[data_coverage_b$sample=="oocyte_2",7])$y,
density(data_coverage_b[data_coverage_b$sample=="oocyte_3",7])$y, density(data_coverage_b[data_coverage_b$sample=="oocyte_4",7])$y,
density(data_coverage_b[data_coverage_b$sample=="oocyte_5",7])$y),
'sample' = c(rep(c("oocyte_1","oocyte_2","oocyte_3","oocyte_4","oocyte_5"),
c(
length(density(data_coverage_b[data_coverage_b$sample=="oocyte_1",7])$y), length(density(data_coverage_b[data_coverage_b$sample=="oocyte_2",7])$y),
length(density(data_coverage_b[data_coverage_b$sample=="oocyte_3",7])$y), length(density(data_coverage_b[data_coverage_b$sample=="oocyte_4",7])$y),
length(density(data_coverage_b[data_coverage_b$sample=="oocyte_5",7])$y))
)
))
plot5<-ggplot(data_coverage_c, aes(x_axis, as.factor(sample))) +
geom_tile(aes(fill = density),colour = "white")+
scale_fill_gradient(low = "#F0F0F0",high = "#000000")+
ggtitle("transcript length < 5,000 nt (6,604 genes)") +
scale_x_continuous(breaks = c(0,100), labels=c("0% 5'", "3' 100%"))+
theme(
panel.background = element_blank(),
plot.background = element_blank(),
panel.grid =element_blank(),
plot.title =element_text(size=font_1, color="black",vjust = 0.4,hjust = 0.5, margin=margin(t =0, r = 0, b = 0, l = 0, unit = "pt")),
axis.text.y =element_text(size=font_1, color="black"),
axis.text.x =element_text(size=font_1, color="black"),
axis.title =element_blank(),
axis.ticks.y =element_blank(),
axis.line =element_blank(),
legend.title =element_text(size=font_2),
legend.text =element_text(size=font_2),
plot.margin =margin(t = 0, r = 0, b = 0, l = 0, unit = "pt"),
legend.key.height=unit(0.2, "cm"),
legend.key.width=unit(0.2, "cm")
)Next, we compiled individual plots to compose the panel observed on the paper. The image may seem distorted on the html file, but is not if saved as a png file.
ggarrange( ggarrange(plot1, plot2,labels = c('a', 'b'), ncol=2, nrow=1, widths = c(2,4),hjust=c(0,0), font.label=list(size = 9, color = "black", face = "plain")),
ggarrange(plot3, labels = c('c'), ncol=1, nrow=1,hjust=0, vjust=1,font.label=list(size = 9, color = "black", face = "plain")),
ggarrange(plot4, ncol=1, nrow=1),
ggarrange(plot5, ncol=1, nrow=1),
nrow=4, heights=c(2,1,1,1.15))
sessionInfo()## R version 3.4.3 (2017-11-30)
## Platform: x86_64-pc-linux-gnu (64-bit)
## Running under: Ubuntu 16.10
##
## Matrix products: default
## BLAS: /usr/lib/atlas-base/atlas/libblas.so.3.0
## LAPACK: /usr/lib/atlas-base/atlas/liblapack.so.3.0
##
## locale:
## [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
## [3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
## [5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
## [7] LC_PAPER=en_US.UTF-8 LC_NAME=C
## [9] LC_ADDRESS=C LC_TELEPHONE=C
## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
##
## attached base packages:
## [1] stats4 parallel stats graphics grDevices utils datasets
## [8] methods base
##
## other attached packages:
## [1] RColorBrewer_1.1-2 biomaRt_2.34.2 kableExtra_0.7.0
## [4] scales_0.5.0 ggpubr_0.1.6 magrittr_1.5
## [7] ggplot2_2.2.1 gplots_3.0.1 stringr_1.2.0
## [10] Rsamtools_1.30.0 Biostrings_2.46.0 XVector_0.18.0
## [13] GenomicRanges_1.30.1 GenomeInfoDb_1.14.0 IRanges_2.12.0
## [16] S4Vectors_0.16.0 BiocGenerics_0.24.0 knitr_1.19
##
## loaded via a namespace (and not attached):
## [1] Rcpp_0.12.15 prettyunits_1.0.2 gtools_3.5.0
## [4] assertthat_0.2.0 rprojroot_1.3-2 digest_0.6.15
## [7] R6_2.2.2 plyr_1.8.4 backports_1.1.2
## [10] RSQLite_2.0 evaluate_0.10.1 highr_0.6
## [13] httr_1.3.1 pillar_1.1.0 zlibbioc_1.24.0
## [16] rlang_0.1.6 progress_1.1.2 curl_3.1
## [19] lazyeval_0.2.1 gdata_2.18.0 blob_1.1.0
## [22] rmarkdown_1.8 labeling_0.3 BiocParallel_1.12.0
## [25] readr_1.1.1 RCurl_1.95-4.10 bit_1.1-12
## [28] munsell_0.4.3 compiler_3.4.3 pkgconfig_2.0.1
## [31] htmltools_0.3.6 tibble_1.4.2 GenomeInfoDbData_1.0.0
## [34] XML_3.98-1.9 viridisLite_0.3.0 dplyr_0.7.4
## [37] bitops_1.0-6 grid_3.4.3 gtable_0.2.0
## [40] DBI_0.7 KernSmooth_2.23-15 stringi_1.1.6
## [43] bindrcpp_0.2 xml2_1.2.0 cowplot_0.9.2
## [46] tools_3.4.3 bit64_0.9-7 Biobase_2.38.0
## [49] glue_1.2.0 purrr_0.2.4 hms_0.4.1
## [52] yaml_2.1.16 AnnotationDbi_1.40.0 colorspace_1.3-2
## [55] caTools_1.17.1 rvest_0.3.2 memoise_1.1.0
## [58] bindr_0.1