Hi everyone,

I have dbGaP metadata for a project that has multiple datatypes (WES, WGS, RNA-seq etc) and I am trying to select only samples corresponding to RNA-seq but I am quite confused looking at a few columns in the metadata. Here are the columns:

> plyr::count(dat[,c('Assay_Type_s','analyte_type_s','molecular_data_type_s')])

  Assay_Type_s analyte_type_s molecular_data_type_s freq
1      RNA-Seq            DNA        <not provided>    1
2      RNA-Seq            RNA        <not provided>  309
3      RNA-Seq            RNA           miRNA (NGS)  150
4      RNA-Seq            RNA         RNA Seq (NGS)  324
5      RNA-Seq            RNA  Targeted Exome (NGS)   32
6      RNA-Seq            RNA     Whole Exome (NGS)    4
7      RNA-Seq            RNA    Whole Genome (NGS)   62

Here, as you can see the Assay Type is RNA-Seq but then you also see Analyte Type as DNA or RNA and Molecular Data Type as either miRNA, RNA-Seq, Targeted, Whole Exome and Whole Genome. How is the relationship determined between these columns and which samples would really be from a RNA-seq experiment?

Thanks