In smaegol/raceseq_scripts: Race-Seq analysis

#knitr::opts_chunk$set(echo = TRUE)
library(dplyr)
library(ggplot2)
library(PMCMRplus)
library(data.table)
library(ggsignif)
#melt_data_localization <- "/home/smaegol/storage/analyses/tail_seq_5/ALL/processing_out/all_samples_2_3_4_5_anal.tsv"
#data_melt<-fread(melt_data_localization)

#data_processed <- process_data(data_melt)

Uridyaltion

Reporter L1 overexpression

Overexpression analyzed first for all mapped reads, irrespective of mapping position (UTR,SV40 signal), only for TUTases and MOV10:

reporterl1_urid <- analyze_uridylation(data_processed,"REPORTERL1_overexp","OVR",include_jitter=T,facet_projects = TRUE,project = c("Tailseq_4","Tailseq_5","Tailseq_2"),conditions=c("CNTRL","TUT7WT","TUT7MT","TUT4WT","TUT4MT","MOV10"))
print(reporterl1_urid$dunn_test)
print(reporterl1_urid$plot)

Analysis of mapping positions revealed differences between experiments:

ovr_mapping_pos <- plot_mapping_positions(data_processed,'REPORTERL1_overexp','OVR',conditions = c("CNTRL"),mapping_position_min = 8900, mapping_position_max = 9200, facet_projects = T)

Therefore, analysis was limited for reads which mapped to the LINE1 UTR, upstream of SV40 signal:

reporterl1_urid <- analyze_uridylation(data_processed,"REPORTERL1_overexp","OVR",include_jitter=T,facet_projects = TRUE,project = c("Tailseq_4","Tailseq_5","Tailseq_2"),conditions=c("CNTRL","TUT7WT","TUT7MT","TUT4WT","TUT4MT","MOV10"),mapping_position_max = 9010)
print(reporterl1_urid$dunn_test)
print(reporterl1_urid$plot)

print("All sqeuencing run together:")
reporterl1_urid <- analyze_uridylation(data_processed,"REPORTERL1_overexp","OVR",include_jitter=T,facet_projects = FALSE,project = c("Tailseq_4","Tailseq_5","Tailseq_2"),conditions=c("CNTRL","TUT7WT","TUT7MT","TUT4WT","TUT4MT","MOV10"),mapping_position_max = 9010)

print("Dunn test statistics:")
print(reporterl1_urid$dunn_test)
print(reporterl1_urid$plot)
#analyze_uridylation()

Run 2 and 4 only

print("Tailseq_2 and Tailseq_4 only:")
reporterl1_urid <- analyze_uridylation(data_processed,"REPORTERL1_overexp","OVR",include_jitter=T,facet_projects = FALSE,project = c("Tailseq_4","Tailseq_2"),conditions=c("CNTRL","TUT7WT","TUT7MT","TUT4WT","TUT4MT","MOV10"),mapping_position_max = 9010)

print("Dunn test statistics:")
print(reporterl1_urid$dunn_test)
print(reporterl1_urid$plot)
#analyze_uridylation()

``` {r tail lengths distribution,message=FALSE,warning=FALSE,echo=FALSE} distrib<-plot_tail_length_distribution(data_processed,"LEAP_AU","LEAP",conditions=c("19A","19A3U"),max_tail_length = 30, min_tail_length=10,facet_conditions = T,AUtail = 'Ataila') print(distrib)

# Reporter KD

#### Analyzed first for all mapped reads, irrespective of mapping position (UTR,SV40 signal), only for TUTases and MOV10:

```r

reporterl1_kd <- analyze_uridylation(processed_data2,"REPORTERL1KD","KD",include_jitter=T,facet_projects = F,project = c("Tailseq_4","Tailseq_5","Tailseq_2"),conditions=c("CNTRLKD","TUT7KD","TUT4KD","MOV10KD","TUT4TUT7KD","TUT4TUT7MOV10KD"))
print(reporterl1_kd$dunn_test)
print(reporterl1_kd$plot)

reporterl1_kd <- analyze_uridylation(processed_data2,"REPORTERL1KD","KD",include_jitter=T,facet_projects = TRUE,project = c("Tailseq_4","Tailseq_5","Tailseq_2"),conditions=c("CNTRLKD","TUT7KD","TUT4KD","MOV10KD","TUT4TUT7KD","TUT4TUT7MOV10KD"))
print(reporterl1_kd$dunn_test)
print(reporterl1_kd$plot)

Analysis of mapping data:

kd_mapping_pos <- plot_mapping_positions(processed_data2,'REPORTERL1KD','KD',conditions = c("CNTRLKD"),mapping_position_min = 8000, mapping_position_max = 9200, facet_projects = T)
print(kd_mapping_pos)

First peak is directly after LINE1 UTR, second is in the BGH adenylation sequence

After filtering To BGH site:

reporterl1_kd <- analyze_uridylation(processed_data2,"REPORTERL1KD","KD",include_jitter=T,facet_projects = F,project = c("Tailseq_4","Tailseq_5","Tailseq_2"),conditions=c("CNTRLKD","TUT7KD","TUT4KD","MOV10KD","TUT4TUT7KD","TUT4TUT7MOV10KD"),mapping_position_min = 8600,mapping_position_max = 8700)
print(reporterl1_kd$dunn_test)
print(reporterl1_kd$plot)

After filtering to before BGH site:

reporterl1_kd <- analyze_uridylation(processed_data2,"REPORTERL1KD","KD",include_jitter=T,facet_projects = F,project = c("Tailseq_4","Tailseq_5","Tailseq_2"),conditions=c("CNTRLKD","TUT7KD","TUT4KD","MOV10KD","TUT4TUT7KD","TUT4TUT7MOV10KD"),mapping_position_max = 8600)
print(reporterl1_kd$dunn_test)
print(reporterl1_kd$plot)

Control of KD (there are 2: HGC and MGC) in the latest sequencing run has lower uridylation than in 1st seq

After filtering to HGC only:

processed_data_no_MGC <- processed_data2 %>% filter((condition=="CNTRLKD" & replicate<7) | condition!="CNTRLKD")
reporterl1_kd <- analyze_uridylation(processed_data_no_MGC,"REPORTERL1KD","KD",include_jitter=T,facet_projects = F,project = c("Tailseq_4","Tailseq_5","Tailseq_2"),conditions=c("CNTRLKD","TUT7KD","TUT4KD","MOV10KD","TUT4TUT7KD","TUT4TUT7MOV10KD"))
print(reporterl1_kd$dunn_test)
print(reporterl1_kd$plot)
reporterl1_kd <- analyze_uridylation(processed_data_no_MGC,"REPORTERL1KD","KD",include_jitter=T,facet_projects = T,project = c("Tailseq_4","Tailseq_5","Tailseq_2"),conditions=c("CNTRLKD","TUT7KD","TUT4KD","MOV10KD","TUT4TUT7KD","TUT4TUT7MOV10KD"))
print(reporterl1_kd$dunn_test)
print(reporterl1_kd$plot)

After filtering to MGC only:

processed_data_no_MGC <- processed_data2 %>% filter((condition=="CNTRLKD" & replicate %in% c(1,2,3,7,8,9)) | condition!="CNTRLKD")
reporterl1_kd <- analyze_uridylation(processed_data_no_MGC,"REPORTERL1KD","KD",include_jitter=T,facet_projects = F,project = c("Tailseq_4","Tailseq_5","Tailseq_2"),conditions=c("CNTRLKD","TUT7KD","TUT4KD","MOV10KD","TUT4TUT7KD","TUT4TUT7MOV10KD"))
print(reporterl1_kd$dunn_test)
print(reporterl1_kd$plot)
reporterl1_kd <- analyze_uridylation(processed_data_no_MGC,"REPORTERL1KD","KD",include_jitter=T,facet_projects = T,project = c("Tailseq_4","Tailseq_5","Tailseq_2"),conditions=c("CNTRLKD","TUT7KD","TUT4KD","MOV10KD","TUT4TUT7KD","TUT4TUT7MOV10KD"))
print(reporterl1_kd$dunn_test)
print(reporterl1_kd$plot)

GAPDH overexpression

reporterl1_urid <- analyze_uridylation(data_processed,"GAPDH","OVR",include_jitter=T,facet_projects = F,project = c("Tailseq_4","Tailseq_5","Tailseq_2"),conditions=c("CNTRL","TUT7WT","TUT7MT","TUT4WT","TUT4MT","MOV10"))
print(reporterl1_urid$dunn_test)
print(reporterl1_urid$plot)
reporterl1_urid <- analyze_uridylation(data_processed,"GAPDH","OVR",include_jitter=T,facet_projects = T,project = c("Tailseq_4","Tailseq_5","Tailseq_3","Tailseq_2"),conditions=c("CNTRL","TUT7WT","TUT7MT","TUT4WT","TUT4MT","MOV10"))
print(reporterl1_urid$dunn_test)
print(reporterl1_urid$plot)

smaegol/raceseq_scripts documentation built on May 28, 2019, 5:40 p.m.