search

pluritest / pluritestCompared

GNU General Public License v3.0
0 stars 0 forks source link

PluriTest for RNA-seq #1

Open emmamonte opened 3 years ago

emmamonte commented 3 years ago

I would like to analyze my RNA-seq samples with PluriTest. It seems that the website no longer accepts RNA-seq data. (Unless I'm just uploading the data incorrectly?) As a result, I would like to try the method via R using the script.R you provide here. However, from your Stem Cell Reports paper in 2017, I understand that in order to use RNA-seq data I first need to generate a TPM matrix and then sum the reads that associate with one of the probe genes. Your paper mentions that you created a modified list of probe genes adapted for RNA-seq, but I couldn't find the list in the supplemental data. I am also unclear how to format the virtual array so that it is accepted for the downstream analyses. I’d really appreciate any advice. Thanks!

ray101023 commented 3 years ago

Hi~would you find out the answer？i also hava the same question. Thanks!

I would like to analyze my RNA-seq samples with PluriTest. It seems that the website no longer accepts RNA-seq data. (Unless I'm just uploading the data incorrectly?) As a result, I would like to try the method via R using the script.R you provide here. However, from your Stem Cell Reports paper in 2017, I understand that in order to use RNA-seq data I first need to generate a TPM matrix and then sum the reads that associate with one of the probe genes. Your paper mentions that you created a modified list of probe genes adapted for RNA-seq, but I couldn't find the list in the supplemental data. I am also unclear how to format the virtual array so that it is accepted for the downstream analyses. I’d really appreciate any advice. Thanks!

emmamonte commented 3 years ago

Hey Ray, So I'm still not sure what the problem was, but several weeks later when I tried again, my RNA-seq files uploaded and ran fine on the PluriTest website. While it'd be nice to do it in R, I just stuck with using the website. I also down-sampled my files. Technically they were under the size limit, but going even lower just made the whole process easier. My "pluripotency score" seemed good. My "novelty score" was a bit off for my samples, potentially due to using RNA-seq data.

ray101023 commented 3 years ago

hahahaha~you are the only one to answer my question.Thank you for your reply!it seems,i am not fighting alone anymore! Let's me talk about my situation. Unfortunately,i have never worked out with the PluriTest website.So,i almost give up to upload my samples.Or maybe i try it again? I search pluritest in Github,then i find a R script from “https://github.com/jhsiao999/pluritest”.Do you have ever use it?I run it succeed and My "pluripotency score" seemed good too,but i got the novelty score in my all samples over 3(if over 1.67,it seems fail),even the sample is ESC(embryonic stem cell).I am so confused about that. And i download some samples from "https://www.hipsci.org/lines/#/lines".The samples from this website will provide pluripotency score and novelty score.But when i run it in this R script,i got different scores. (it also upload fail in pluritest website >.< ) i don't know the problem,maybe the pluritest have been upgreaded or i input the wrong data in R script(cpm value matrix,the R script suggest to do it). so,i don't know what to do now. Finally,i want create a local pluritest with my own dataset,but i stuck in the command line "coef<-c(-1.267095e+02 ,4.567437e-03 , 4.377068e-03 ,1.043193e-03)",it seems only fit the dataset in W15 which it provide.(or maybe i think wrong,i don't know).

.< if you have any advices,you can contact my e-mail:ray101023@163.com,or answer it here. Thanks a lot again!