Differentially expressed genes with sleuth- RStudio

Question

[amina.alhalwany]

Hello I need a help in analyzing deferentially expressed genes with I have a Kallisto data with 6 folders for RNA seq. and an identification file for info. as the following:

run_accession   experiment_accession    Index   condition   sample

pC1-a   pC1-6a  Rip-GFP/Col-a   control Xiao-Ning
pC2-b   pC1-6b  Rip-GFP/Col-b   control Xiao-Ning
pC3-c   pC1-6c  Rip-GFP/Col-c   control Xiao-Ning
p9-1    p9-1-6g Rip-OX9-1   p9  Xiao-Ning
p9-2    p9-2-6h Rip-OX9-2   p9  Xiao-Ning
p9-3    p9-3-6j Rip-OX9-3   p9  Xiao-Ning

and my task is analyzing them with the sleuth on R studio I have a certain script as the following:

source("http://bioconductor.org/biocLite.R")
biocLite("rhdf5")
biocLite("biomaRt")

biocLite("devtools") # only if devtools not yet installed
install.packages('devtools')
devtools::install_github('pachterlab/sleuth')
install.packages('shiny')

library("sleuth")

base_dir <- "C:/Users/asus/Desktop/data"

sample_id <- dir(file.path(base_dir,"results"))

kal_dirs <- sapply(sample_id, function(id) file.path(base_dir, "results", id, "kallisto"))

s2c <- read.table(file.path(base_dir, "hiseq_info_rip_p9.txt" ), header = TRUE, stringsAsFactors=FALSE)
s2c <- dplyr::select(s2c, sample = run_accession, condition)
s2c
s2c <- dplyr::mutate(s2c, path = kal_dirs)
print(s2c)
so <- sleuth_prep(s2c, ~ condition)
so <- sleuth_fit(so)
so <- sleuth_wt(so, 'conditionp1')
models(so)
sleuth_live(so)
results_table <- sleuth_results(so, 'conditionp9')

and the script is working but I am facing a problem as the following:

reading in kallisto results
......
Error in check_kal_pack(kal_list) :
Inconsistent number of transcripts. Please make sure you used the same index everywhere.

so <- sleuth_fit(so)
Error in is(obj, "sleuth") : object 'so' not found
so <- sleuth_wt(so, 'control    Xiao-Ning')
Error in is(obj, "sleuth") : object 'so' not found
models(so)
Error in models(so) : object 'so' not found
sleuth_live(so)
Error in is(obj, "sleuth") : object 'so' not found
results_table <- sleuth_results(so, 'Xiao-Ning')
Error in is(obj, "sleuth") : object 'so' not found

---

I would appreciate any small details that could help, thank you!

[Devon Ryan]

The key is "Inconsistent number of transcripts", which suggests that you quantified against different things. Fix that.

[amina.alhalwany]

can you help me fixing it ?

[h.mon]

Just run kallisto against the same reference transcriptome for every sample. If you already did that, you probably have some truncated file.