diff --git a/assays/Salmon_quantification/isa.assay.xlsx b/assays/Salmon_quantification/isa.assay.xlsx
index 036a309463037e7836c3b4332d4e5b02589575e9..f9a63ed6a42fc57f261ae548b713107ec640e5f2 100644
Binary files a/assays/Salmon_quantification/isa.assay.xlsx and b/assays/Salmon_quantification/isa.assay.xlsx differ
diff --git a/assays/Salmon_quantification/protocols/TranscriptQuantification.md b/assays/Salmon_quantification/protocols/TranscriptQuantification.md
new file mode 100644
index 0000000000000000000000000000000000000000..e1334ae339503e2fd2d7084fd23f5ebd9cd91390
--- /dev/null
+++ b/assays/Salmon_quantification/protocols/TranscriptQuantification.md
@@ -0,0 +1,3 @@
+## Transcript Quantification with Salmon
+
+To quantify transcripts, all cleaned reads were mapped to the BaRT1 reference (Rapazote-Flores et al., 2019) using Salmon (v. 0.14.1) (Patro et al., 2017), and the mapping rates were estimated with an average of ∼93%. We kept transcripts with a minimum of 1 CPM (counts per million) in at least three samples. Normalization of TPM (transcript per million) was conducted using the ‘ThreeDRNAseq’ R package (v2.0.1) (Guo et al., 2021). For TPM values, see Supplementary Dataset 2 and 3. As the samples were collected at different DAEs and RNA extraction was conducted in several batches, we removed the batch effect in the gene count by using the ThreeDRNAseq’ R package (v2.0.1) (Guo et al., 2021).
\ No newline at end of file