cov.200bp.wind

Map reads to bins

The binned genome file is called: genome_200_wind_fix2.saf, it is in my genome annotation directory.

#!/bin/bash

#SBATCH --job-name=cov200
#SBATCH --time=8:00:00
#SBATCH --output=cov200.out
#SBATCH --error=cov200.err
#SBATCH --partition=broadwl
#SBATCH --mem=20G
#SBATCH --mail-type=END

module load Anaconda3  

source activate three-prime-env

#input is a bam 
sample=$1


describer=$(echo ${sample} | sed -e 's/.*\YL-SP-//' | sed -e "s/-sort.bam$//")



featureCounts -T 5 -a /project2/gilad/briana/genome_anotation_data/an.int.genome_200_strandspec.saf -F 'SAF' -o /project2/gilad/briana/threeprimeseq/data/cov_200/${describer}_FC200.cov.bed $1

I will need to create a wrapper to run this for all of the files.

#!/bin/bash

#SBATCH --job-name=w_cov200
#SBATCH --time=8:00:00
#SBATCH --output=w_cov200.out
#SBATCH --error=w_cov200.err
#SBATCH --partition=broadwl
#SBATCH --mem=8G
#SBATCH --mail-type=END


for i in $(ls /project2/gilad/briana/threeprimeseq/data/sort/*.bam); do
            sbatch cov200.sh $i 
        done

Current analysis is not stand specific. I need to make windows for the negative strand. To do this I need to copy the genome_200_wind_fix2.saf file but with geneIDs starting with the last number of the file and with a - for the strand. The last window number is 15685849. I will have to start from 15685850.

In general I will use awk to create the file. The last number is 31371698 because that is 2 * the number of bins in the genome. I w

#i will delete the top line at the end
seq 15685849 31371698 > neg.bin.num.txt

 cut -f1 neg.bin.num.txt | paste - genome_200_wind_fix2.saf | awk  '{ if (NR>1) print $1 "\t" $3 "\t" $4 "\t" $5 "\t" "-"}' >  genome_200_wind_fix2.negstrand.saf 

#cat files together  

cat genome_200_wind_fix2.saf  genome_200_wind_fix2.negstrand.saf  > genome_200_strandsspec_wind.saf
 

I can use this to get coverage in all of the windows with strand specificity. I will call this script ss_cov200.sh

#!/bin/bash

#SBATCH --job-name=sscov200
#SBATCH --time=8:00:00
#SBATCH --output=sscov200.out
#SBATCH --error=sscov200.err
#SBATCH --partition=broadwl
#SBATCH --mem=20G
#SBATCH --mail-type=END

module load Anaconda3  

source activate three-prime-env

#input is a bam 
sample=$1


describer=$(echo ${sample} | sed -e 's/.*\YL-SP-//' | sed -e "s/-sort.bam$//")



featureCounts -T 5 -s 1 -O --fraction -a /project2/gilad/briana/genome_anotation_data/genome_200_strandsspec_wind.saf -F 'SAF' -o /project2/gilad/briana/threeprimeseq/data/ss_cov200/${describer}_ssFC200.cov.bed $1

Try this with. /project2/gilad/briana/threeprimeseq/data/sort/YL-SP-18486-N_S10_R1_001-sort.bam

I will update my wrapper to use this script.

The current script does not allow reads that map to multiple bins. We expect then so I will update the featureCounts code to account for this.

-O allows multi mapping -fraction will put a fraction of the read in each bin

The next step is to add genes annotations to each bin. I will do this with bedtools closest on my window file.

gene file: /project2/gilad/briana/genome_anotation_data/gencode.v19.annotation.proteincodinggene.sort.bed

I want to keep the windows with gene and add the name of the gene they are in.

a= windows b= genes

force stranded= -s

I need to make the window file a sorted bed file. It should be the chr number without the ‘chr’ tag, start, end, bin number, “.”, strand.

awk '{if (NR>1) print $2 "\t" $3 "\t" $4 "\t" $1 "\t" "." "\t" $5}' genome_200_strandsspec_wind.saf  | sed 's/^chr//' | sort -k1,1 -k2,2n > genome_200_strandspec.bed

#!/bin/bash

#SBATCH --job-name=annotate_wind
#SBATCH --time=8:00:00
#SBATCH --output=an_wind.out
#SBATCH --error=an_wind.err
#SBATCH --partition=broadwl
#SBATCH --mem=30G
#SBATCH --mail-type=END

module load Anaconda3

source activate three-prime-env


bedtools closest -s -a genome_200_strandspec.bed -b gencode.v19.annotation.proteincodinggene.sort.bed > annotated.genome_200_strandspec.bed

Now i can use intersect to only keep the windows that interdect that protien coding genes.

#!/bin/bash

#SBATCH --job-name=int_wind
#SBATCH --time=8:00:00
#SBATCH --output=int_wind.out
#SBATCH --error=int_wind.err
#SBATCH --partition=broadwl
#SBATCH --mem=30G
#SBATCH --mail-type=END

module load Anaconda3

source activate three-prime-env


bedtools intersect -wa -sorted -s -a annotated.genome_200_strandspec.bed -b gencode.v19.annotation.proteincodinggene.sort.bed > annotated.int.genome_200_strandspec.bed

awk '{print $1 "\t" $2 "\t" $3 "\t" $4 "\t" $5 "\t" $6 "\t" $10}'  annotated.int.genome_200_strandspec.bed >  an.int.genome_200_strandspec.bed

I went from 31590487 to 7371747 windows. I need to make this into a saf file and the name of the window will be the number.gene

awk '{print $4"."$7 "\t" $1 "\t" $2 "\t" $3 "\t" $6}'  an.int.genome_200_strandspec.bed > an.int.genome_200_strandspec.saf

#go into the file with vi and add header

Now I can change my feature counts script to use this file instead.

I need to get rid of the lines with 2 genes overlapping in the bin. I will do this by removing the lines with a :.

for i in $(ls *.bed); do
      cat $i | grep -v -e ";" > ../ss_cov200_no_overlap/$i
  done

The next step is to bind all of these files. This file will have all 6323877 windows as the rows and columns for each of the 32 files

less 18486-N_S10_R1_001_ssFC200.cov.bed  | cut -f1-6 > tmp 
for i in ./*cov.bed; do
echo "$i"
less ${i} | cut -f7 >col
paste tmp col> tmp2; mv tmp2 tmp; rm col; done

mv tmp ssFC200.cov.bed

This in now ready to move to R an work with it here.

Assess bin coverage

library(workflowr)

Loading required package: rmarkdown

This is workflowr version 1.0.1
Run ?workflowr for help getting started

library(ggplot2)
library(dplyr)

Attaching package: 'dplyr'

The following objects are masked from 'package:stats':

    filter, lag

The following objects are masked from 'package:base':

    intersect, setdiff, setequal, union

library(tidyr)
library(edgeR)

Warning: package 'edgeR' was built under R version 3.4.3

Loading required package: limma

Warning: package 'limma' was built under R version 3.4.3

library(reshape2)

Warning: package 'reshape2' was built under R version 3.4.3

Attaching package: 'reshape2'

The following object is masked from 'package:tidyr':

    smiths

names=c("N_18486","T_18486","N_18497","T_18497","N_18500","T_18500","N_18505",'T_18505',"N_18508","T_18508","N_18853","T_18853","N_18870","T_18870","N_19128","T_19128","N_19141","T_19141","N_19193","T_19193","N_19209","T_19209","N_19223","N_19225","T_19225","T_19223","N_19238","T_19238","N_19239","T_19239","N_19257","T_19257")

cov_all=read.table("../data/ssFC200.cov.bed", header = T, stringsAsFactors = FALSE)
#remember name switch!
names=c("Geneid","Chr", "Start", "End", "Strand", "Length", "N_18486","T_18486","N_18497","T_18497","N_18500","T_18500","N_18505",'T_18505',"N_18508","T_18508","N_18853","T_18853","N_18870","T_18870","N_19128","T_19128","N_19141","T_19141","N_19193","T_19193","N_19209","T_19209","N_19223","N_19225","T_19225","T_19223","N_19238","T_19238","N_19239","T_19239","N_19257","T_19257")

colnames(cov_all)= names

Plot the density of the log of the counts.

cov_nums_only=cov_all[,7:38]
cov_nums_only_log=log10(cov_nums_only)
plotDensities(cov_nums_only_log,legend = "bottomright", main="bin log 10 counts")

Now I want to filter for bins that have 0 reads in >16 samples.

keep.exprs=rowSums(cov_nums_only>0) >= 16

cov_all_filt=cov_all[keep.exprs,]
bin.genes=cov_all_filt[,1]

I will now look at the densities.

cov_all_filt_log=log10(cov_all_filt[,7:38] + 1) 

plotDensities(cov_all_filt_log,legend = "bottomright", main="Filtered bin log10 +1 counts")

I want to make boxplots for each of these lines. I should tidy the data with a column for total or nuclear.

sample=c("N_18486","T_18486","N_18497","T_18497","N_18500","T_18500","N_18505",'T_18505',"N_18508","T_18508","N_18853","T_18853","N_18870","T_18870","N_19128","T_19128","N_19141","T_19141","N_19193","T_19193","N_19209","T_19209","N_19223","N_19225","T_19225","T_19223","N_19238","T_19238","N_19239","T_19239","N_19257","T_19257")
fraction=c("N","T","N","T","N","T","N",'T',"N","T","N","T","N","T","N","T","N","T","N","T","N","T","N","N","T","T","N","T","N","T","N","T")

cov_all_filt_log_gen=cbind(bin.genes,cov_all_filt_log)

cov_all_tidy= cov_all_filt_log_gen%>% gather(sample, value, -bin.genes)


#add fraction column

cov_all_tidy_frac=cov_all_tidy %>% mutate(fraction=ifelse(grepl("T",sample), "total", "nuclear")) %>% mutate(line=substr(sample,3,7))

Make a heatmap:

bin_count=ggplot(cov_all_tidy_frac, aes(x = line, y=value,fill=fraction )) + geom_boxplot(position="dodge") + labs(y="log10 count + 1", title="Bins in nuclear fractions have larger counts " )
bin_count

#ggsave("../output/plots/bin_counts_by_line.png", bin_count)

cov_all_filt_genes=separate(data = cov_all_filt, col = Geneid, into = c("bin", "gene"), sep = ".E") 
cov_all_filt_genes$gene= paste( "E",  cov_all_filt_genes$gene, sep="" )
cov_all_filt_num=cov_all_filt_genes[,8:39]
non_zero=colSums(cov_all_filt_num != 0)

#make a data frame to plot this

non_zero_df=data.frame(non_zero)
non_zero_df= non_zero_df %>% mutate(fraction=ifelse(grepl("T",rownames(non_zero_df)), "total", "nuclear")) %>% mutate(line=substr(rownames(non_zero_df),3,7))


non_zero_plot=ggplot(non_zero_df, aes(x = line, y=non_zero, fill=fraction )) + geom_bar(position="dodge",stat="identity") + labs(y="Non zero bins", title="Number of bins with reads after filtering")
non_zero_plot

#ggsave("../output/plots/non_zero_bins.png", non_zero_plot)

cov_all_filt_small=cbind(cov_all_filt_genes[,1:2],cov_all_filt_genes[,8:39])
cov_all_filt_pergene=cov_all_filt_small %>% gather(sample, value, -gene, -bin)  %>% group_by(gene, sample) %>% summarise(non_zero=sum(value!=0)/n())%>% mutate(fraction=ifelse(grepl("T",sample), "total", "nuclear")) %>% mutate(line=substr(sample,3,7))

Prepare data for leafcutter:

For leafcutter I need the data to look like:

chr1:APA1:gene_name count.ind1 count.ind2

chr1:APA2:gene_name ind1 count.ind2

I will separate the fractions into 2 data frames them filter each by bins with at least 5 counts in 1/3 of the individuals.

#counts only
cov_nuc=cov_all %>% select(contains("N_"))
#with annotations
cov_nuc_anno=cbind(cov_all[,1:6], cov_nuc)

keep.nuc= rowSums(cov_nuc>=5) >= 5

#annotated and filtered nuclear
cov_nuc_anno_filt=cov_nuc_anno[keep.nuc,]

Run the same filter for the total fraction.

#counts total only  
cov_tot=cov_all %>% select(contains("T_"))
#with annotaiton
cov_tot_anno=cbind(cov_all[,1:6], cov_tot)

keep_tot=rowSums(cov_tot>=5)>=5

#annotated and filtered total

cov_tot_anno_filt=cov_tot_anno[keep_tot,]

Now I need to change the annoation to be chrom:apa#:gene. To do this I need to know how many bins for each bin are in the file. I can use groupby and summarize to do this.

#nuclear genes

genes_nuc= cov_nuc_anno_filt %>% separate(col = Geneid, into = c("bin", "gene"), sep = ".E") %>% group_by(gene) %>% select(gene) %>% tally()
genes_nuc$gene= paste( "E",  genes_nuc$gene, sep="" )



#total genes


genes_tot=cov_tot_anno_filt %>% separate(col=Geneid, into=c("bin","gene"), sep=".E") %>% group_by(gene) %>% select(gene) %>% tally()


genes_tot$gene=paste("E", genes_tot$gene, sep="")

Now I need a way to make a vector with APA# counting up for the number of bins in each gene.

#nuclear APA
apa_nuc=c()
for (row in 1:nrow(genes_nuc)){
  x=1
  i=1
  while(i <= as.numeric(genes_nuc[row,2])){
    apa_nuc= c(apa_nuc, paste("APA", x, sep = ""))
    x= x + 1
    i= i + 1
  }
}

#total APA
apa_tot=c()
for(row in 1:nrow(genes_tot)){
  x=1
  i=1
  while(i<= as.numeric(genes_tot[row,2])){
    apa_tot=c(apa_tot, paste("APA", x, sep=""))
    x= x + 1
    i= i + 1
  }
}

The apa_tot and apa_nuc vector now number the bins with reads for each gene. I can use this to make the table.

cov_tot_anno_filt_group= cov_tot_anno_filt %>%separate(col=Geneid, into=c("bin","gene"), sep=".E") %>% group_by(gene) %>% arrange(gene)
cov_tot_anno_filt_group$gene=  paste( "E",  cov_tot_anno_filt_group$gene, sep="" )
total_anno=paste(cov_tot_anno_filt_group$Chr, apa_tot, cov_tot_anno_filt_group$gene, sep=":")
total_leaf=cbind(total_anno, cov_tot_anno_filt_group[,8:22]) 

To this for nuclear:

cov_nuc_anno_filt_group = cov_nuc_anno_filt %>% separate(col=Geneid, into=c("bin","gene"), sep=".E") %>% group_by(gene) %>% arrange(gene)
cov_nuc_anno_filt_group$gene=paste("E",cov_nuc_anno_filt_group$gene, sep="")
nuc_anno=paste(cov_nuc_anno_filt_group$Chr,apa_nuc,cov_nuc_anno_filt_group$gene, sep=":")
nuc_leaf=cbind(nuc_anno,cov_nuc_anno_filt_group[,8:22])

Write both of these tables out:

#write.csv(nuc_leaf, file="../data/leafcutter/nuc_apa_200wind.csv",row.names = FALSE, quote = FALSE)
#write.csv(total_leaf, file="../data/leafcutter/tot_apa_200wind.csv",row.names = FALSE, quote = FALSE)

I actually want both of the samples together and I want the name of the site to be representative of the bin. To do this I can just use the start and end from the bed file. I also will create one table then filter at the end.

cov_all_anno=cov_all %>% separate(col=Geneid, into=c("bin","gene"), sep=".E") 
cov_all_anno$gene=  paste( "E",  cov_all_anno$gene, sep="" )
bin_loc=paste(cov_all_anno$Start, cov_all_anno$End, cov_all_anno$Strand,sep=".")

leaf_all_anno=paste(cov_all_anno$Chr,bin_loc, cov_all_anno$gene, sep=":")

leaf_all=cbind(leaf_all_anno,cov_all_anno[,8:39])

Filter this with an or statement.

leaf_all_nuc= leaf_all %>% select(contains("N_"))
keep.nuc.leaf=rowSums(leaf_all_nuc>=5) >= 5
leaf_nuc_filt=leaf_all[keep.nuc.leaf,]


leaf_all_tot= leaf_all %>% select(contains("T_"))
keep.tot.leaf=rowSums(leaf_all_tot>=5) >= 5
leaf_tot_filt=leaf_all[keep.tot.leaf,]

leaf_all_filt=union(leaf_nuc_filt,leaf_tot_filt)

Write this out:

#write.table(leaf_all_filt, file="../data/leafcutter/all_leaf_200wind.csv", row.names = FALSE, quote = FALSE, sep=" ")

I need to get rid of the first col name. I can do this in vim on the command line.

Session information

sessionInfo()

R version 3.4.2 (2017-09-28)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS Sierra 10.12.6

Matrix products: default
BLAS: /Library/Frameworks/R.framework/Versions/3.4/Resources/lib/libRblas.0.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.4/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] bindrcpp_0.2    reshape2_1.4.3  edgeR_3.20.9    limma_3.34.9   
[5] tidyr_0.7.2     dplyr_0.7.4     ggplot2_2.2.1   workflowr_1.0.1
[9] rmarkdown_1.8.5

loaded via a namespace (and not attached):
 [1] Rcpp_0.12.15      compiler_3.4.2    pillar_1.1.0     
 [4] git2r_0.21.0      plyr_1.8.4        bindr_0.1        
 [7] R.methodsS3_1.7.1 R.utils_2.6.0     tools_3.4.2      
[10] digest_0.6.14     lattice_0.20-35   evaluate_0.10.1  
[13] tibble_1.4.2      gtable_0.2.0      pkgconfig_2.0.1  
[16] rlang_0.1.6       yaml_2.1.16       stringr_1.2.0    
[19] knitr_1.18        tidyselect_0.2.3  locfit_1.5-9.1   
[22] rprojroot_1.3-2   grid_3.4.2        glue_1.2.0       
[25] R6_2.2.2          purrr_0.2.4       magrittr_1.5     
[28] whisker_0.3-2     backports_1.1.2   scales_0.5.0     
[31] htmltools_0.3.6   assertthat_0.2.0  colorspace_1.3-2 
[34] labeling_0.3      stringi_1.1.6     lazyeval_0.2.1   
[37] munsell_0.4.3     R.oo_1.22.0