diff --git a/studies/AccessionSelection/README.md b/studies/AccessionSelection/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/studies/AccessionSelection/isa.study.xlsx b/studies/AccessionSelection/isa.study.xlsx
new file mode 100644
index 0000000000000000000000000000000000000000..49ae4d1998d54435e6ca26a75d65e7ddc1ca4567
Binary files /dev/null and b/studies/AccessionSelection/isa.study.xlsx differ
diff --git a/studies/AccessionSelection/protocols/.gitkeep b/studies/AccessionSelection/protocols/.gitkeep
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/studies/AccessionSelection/protocols/AccessionSelectionProtocol.md b/studies/AccessionSelection/protocols/AccessionSelectionProtocol.md
new file mode 100644
index 0000000000000000000000000000000000000000..1f646b8ca084e4cea7126b595c8c1933f5851f89
--- /dev/null
+++ b/studies/AccessionSelection/protocols/AccessionSelectionProtocol.md
@@ -0,0 +1,3 @@
+## Accession selection
+
+In order to establish clusters of *A. thaliana* accessions with high versus low S content in their leaves, we re-analyzed ionomics data from Baxter et al. (2007) and Campos et al. (2021), with 174 accessions overlapping in the two datasets (Baxter et al., 2007; Campos et al., 2021). The two datasets of Baxter et al. (2007) and Campos et al. (2021) have different ranges of S concentrations, and individual accessions differ in absolute value but are usually in a similar percentile or relative range, with some exceptions. Therefore, we aimed to use only the overlapping accessions and to use both datasets to decide which accessions to choose. From these 174 overlapping accessions, a principal component analysis (PCA) was performed on the total of 37 ion data (Supplementary Table S1) to establish groupings of accessions with similar leaf S content. Based on the availability of these accessions, we selected 20 accessions with low, medium, and high sulfate levels for further investigation (Supplementary Fig. S1; Supplementary Table S1).
\ No newline at end of file
diff --git a/studies/AccessionSelection/resources/.gitkeep b/studies/AccessionSelection/resources/.gitkeep
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391