CONCLUSIONS
After discussing and analyzing our results, here we present the conclusions of this project:
Firstly, we can conclude that we have identified 59 selenoproteins in Oncorhynchus mykiss : Sel15_1, Sel15_2, SELENOE_1, SELENOE_3, GPx1a_1, GPx1a_2, GPx1b_1, GPx1b_2, GPx2, GPx3a_1, GPx3a_2, GPx3b, GPx4a, GPx4b_1, GPx4b_2, DIO1, DIO2a_1, DIO2a_2, DIO2a_3, DIO3a_1 DIO3a_2, SEPHS2, SELENOH_1, SELENOH_2, SELENOI_1, SELENOI_2, SELENOJ_2, SELENOK, SELENOL, SELENOM_1, SELENOM_2, SELENON, SELENOO1-2_1, SELENOO1-2_2, SELENOO1-2_3, SELENOP_1.1, SELENOP_1.2, SELENOP_2.1, SELENOP_2.2, MSRB1a_1, MSRB1a_2, MSRB1b, MSRB2_1, MSRB2_2,, SELENOS, SELENOT1_1, SELENOT1_2, SELENOT1b, SELENOT2, SELENOU1a_1, SELENOU1a_2, SELENOU1a_3, SELENOW_1.1, SELENOW_1.2, SELENOW_3, TXNRD2_1, TXNRD2_2, TXNRD3_1, and TXNRD3_2.
Regarding Cys-containing homologous proteins, it has been concluded that Oncorhynchus mykiss's genome contains 12 of them: GPx7, GPx8, SELENOJ_1, MsrA_1, MsrA_2.1, MsrA_2.2, MSRB2_1, MSRB2_2, MSRB3_1, MSRB3_2, SELENOU2 and SELENOU3.
It has also been concluded that Oncorhynchus mykiss contains these 11 selenoprotein machinery genes: SecS_1, SecS_2, SBP2, PSTK, eEFsec, SEPHS1_1, SEPHS1_2, SEPSH2, SECp43_1, SECp43_2.1, SECp43_2.2.
The machinery protein PSTK has been identified but the predicted gene is dubious, therefore, its presence in Oncorhynchus mykiss can not be confirmed. Moreover, it is worth pointing out the absence of DIO3b, a seleneprotein specific from bony fishes.
Particularly, some of the proteins predicted in this project do not start with a methionine: this is due to the fact that the some of the original proteins from zebrafish are not well annotated in SelenoDB and do not start by a methionine residue. For this reason, is possible that some of the predictions of this project comprise only fragments of these proteins. In order to solve this, other databases could be used to align Oncorhynchus mykiss's genome with these specific queries.
Another issue to comment is the SECIS prediction. SECIS are secondary mRNA structures and SEBLASTIAN predicts every similar structure found in the DNA sequence introduced. One of the problems encountered is that in some of the predicted selenoproteins no SECIS were predicted while for some machinery proteins multiple SECIS were identified. The later could be linked to the high sensitivity of the algorithm. Even though, in this context we can not assess if this unexpected absence or presence is due to prediction problems or to evolutionary associated processes.
Overall, as multiple selenoproteins have been predicted in Oncorhynchus mykiss's genome, this information could be useful for further studies about selenoproteomes in bony fishes.
