Quantify the distribution of RNAs from cellular fractions.
This package employs a bayesian statistical model to quantify the distribution of mRNA
Users need to provide a count matrix and a samplesheet file. Those files need to meet the conditions below:
1. Both the count matrix and samplesheet file need to be in the ".tsv" format
2. The count matrix should have a column of transcript_IDs then a column of counts for every Sample_ID. It is highly recommended to first filter this for verified ORFs. Counts can be integers or fraction estimates (e.g. est_counts from kallisto).
| transcript_ID | F01~snake_230331 | F02~snake_230331 |
|---|---|---|
| YAL068C | 1.08 | 3.41 |
| YAL067C | 17 | 15 |
| YAL064W | 16 | 24 |
**3. In samplesheet file, there should be at least four columns "Sample_ID","Condition","Rep",and "Fraction". Every Sample_ID must be unique. For every Condition and Rep, there should be a Sample_ID corresponding to the three fractions: Total, Supernatant, and Pellet **
| Sample_ID | Condition | Rep | Fraction |
|---|---|---|---|
| F01~snake_230331 | hairpinReporters |
yHG010 | Total |
| F02~snake_230331 | hairpinReporters |
yHG005 | Total |
4. For every line of comment, there should be a tag "#" at the begining of the text
We recommend you to use package "devtools" for dowloading this package from GitHub. Please refer devtools installation instructions for more information.
install.packages("devtools")
devtools::install_github("jabard89/SedSeqQuant")
Then load SedSeqQuant as a standard package:
library(SedSeqQuant)
count_data <- load_counts(count_file=file.path("count_matrix.tsv.gz"),
samplesheet = file.path("samplesheet.tsv"))
pd_noreps <- prepare_data_noreps(count_data,min_counts=20)
stanfit_noreps <- model_fit_noreps(pd_noreps)
sum_noreps <- get_stan_summary_noreps(stanfit_noreps,pd_noreps)
write_stan_summary_noreps(sum_noreps,file.path(wrk.dir,"output/no_reps"))
pd_withreps <- prepare_data_with_reps(count_data,min_counts=20)
stanfit_withreps <- model_fit_reps(pd_withreps,iter=8000,chains=4)
sum_withreps <- get_stan_summary_reps(stanfit_withreps,pd_withreps)
write_stan_summary_with_reps(sum_withreps,file.path(wrk.dir,"output/with_reps"))