We studied the structuredness ensemble of transcriptome of Siberian larch. The clusters in 64-dimensional space were identified with $K$-means technique, where the objects to be clusterized are the different fragments of the genome. A tetrahedron like structure in distribution of these fragments was found. Chargaff's discrepancy measure was determined for each class, as well as that latter between the classes. It reveals a relative similitude of the classes. The results have been compared to those obtained for specific transcriptome of each tissue. Also, a surrogate transcriptome has been developed comprising the contigs assembled for specific tissues; that latter has been compared with the real total transcriptome, and significant difference has been observed.