Fig 1.
Relationship between piRNA biogenesis (transcription, processing and function) and measured features.
(i) piRNA clusters can be transcribed if particular methylated histone (fly) or A-Myb promoter (mouse) is nearby; (ii) G-quadruplexes could have a role in piRNA processing and (iii) both first and tenth piRNA bases (respectively U and A) represent an important binding zone for Argonaute proteins, participating in a ping-pong cycle where the piRNA sequences bind with transposons.
Table 1.
piRNA’s biological features over species.
Fig 2.
The different kernel classes defined in IpiRId and their hierarchical organisation.
Table 2.
IpiRId’s kernels instantiation.
(D: distance; L: minimal length).
Table 3.
The downloaded data used in our integrative approach for piRNAs identificatiton across species.
Table 4.
5-fold cross-validation results of IpiRId and other existing tools according to: Accuracy (Acc), Sensitivity (Se), Specificity (Sp), Precision (Pre) and F1 score (F1).
Fig 3.
ROC space and plots of the 5-fold cross-validation results of IpiRId and other tools across species, with fixed parameters.
Fig 4.
IpiRId prediction results on piRNA and pseudo-piRNA sequences across species.
Fig 5.
IpiRId’s features pertinence across species: Mouse, Human and Fly.