Recently I did RNA seq on mycobacterium. I got the TPM data from one of our colleagues in the Bioinformatics department because they helped us to analyze the raw data.
Then, I'm interested in making a volcano plot from the data. Do you think it's possible to get a p-value from TPM data?
You might want to look into these threads:
https://support.bioconductor.org/p/98820/#98875
https://www.biostars.org/p/9470568/
Also, get your hands on the raw data if you can. The raw counts are required for DEG analysis rather than normalized TPM values.
TPM is a metric that normalize counts by gene length and sequencing depth and it's useful to compare gene expression within the same sample. I would try to get the raw counts and start a fresh differential expression analysis.
Hi Desak Nyoman Surya Suameitria Dewi, RNA-seq data should be processed with dedicated statistics, such as DESeq2 - Bioconductor or edgeR - Bioconductor. If you use p-values calculated with those statistics, you can create volcano plots. The reason why you cannot use more typical statistics, such as ANOVA, to get the p-values, is that RNA -seq data does not have a normal distribution, and many common inferential statistics probes require data with normal distribution.
Best !!
AN
Abdullah Al Nahid Anuraj Nayarisseri Pedro de los Reyes Rodríguez Thank you so much for all your answer. I got it.
TPM is a gene count normalization unit (read more here https://www.reneshbedre.com/blog/expression_units.html). You need to perform differential expression analysis using a tool such as DESeq2 (read more here https://www.reneshbedre.com/blog/deseq2.html) to get p value and to create a volcano plot (https://www.reneshbedre.com/blog/volcano.html)
- Badges
- Science topic
- Similar topics
- Agriculture
- Horticulture
- Seeds