Following steps after Functional Annotation of P. Generosa transcriptome part 3

overlapping genes counts by tissue type

load count data from kallisto output(s)

countMatrix<-read.table(file="https://raw.githubusercontent.com/sr320/nb-2022/main/P_generosa/analyses/kallisto-0207.isoform.counts.matrix", header=TRUE, sep = '\t')
names(countMatrix)[1]<-"target_id"
head(countMatrix)

use binary data to make Venn Diagram

library(tidyr)
library(dplyr)
simple_data<-countMatrix
long_data<- reshape(data=simple_data, idvar="target_id", varying=c("heart", "gonad", "ctenidia", "juv_sl", "juv_amb", "larvae"), v.name=c("tpm"), times=c("heart", "gonad", "ctenidia", "juv_sl", "juv_amb", "larvae"), new.row.names=1:8183754, direction="long")
names(long_data)[2]<-"tissue"
summary(long_data)
long_data$binary=ifelse(long_data$tpm > 0, 1, 0)
long_data_zeros<-long_data[apply(long_data, 1, function(row) all(row !=0)),]
df<-long_data_zeros[-c(3,4)] #tissue type and gene names only
table_no_zeros<-as.data.frame(crossprod(table(df)))
#write.csv(table_no_zeros, file="/Users/oliviacattau/Documents/GitHub/code/characterize_larval_transciptome/pgenerosa_table_no_zeros.tab")
simple_data_cleaned<-simple_data[apply(simple_data, 1, function(row) all(row !=0)),]
genelist<-long_data_zeros %>% group_by(tissue)
genelist %>% summarise("total reads (n)"=sum(binary),  "mean tpm" = mean(tpm), "Q4 tpm" = quantile(tpm, c(0.75)))

link to crossprot table

link to dplyr table