Tips:

The input file should be a FASTA format file containing sequences of plant mitochondrial protein-coding genes. The file can contain a single sequence or multiple sequences. Once the file is uploaded, this module will use the model trained by Deepredmt to predict the probability of C-to-U RNA editing occurring at each cytosine in the sequences and return these probability values.

If you use Deepred-Mt, please remember to cite: Edera AA, Small I, Milone DH, Sanchez-Puerta MV: Deepred-Mt: Deep representation learning for predicting C-to-U RNA editing in plant mitochondria. Comput Biol Med 2021, 136:104682.

MitoRED is an optimized model based on Deepred-Mt. Building upon the original Deepred-Mt dataset, MitoRED incorporates additional RNA editing sites discovered from RNA-seq data of 25 angiosperm species. It utilizes longer input sequences (centered on the target cytosine with 40 bp flanking regions on each side) as model input, significantly reducing the false positive rate. MitoRED improves the accuracy of predicting conserved C-to-U RNA editing sites, but it is not suitable for identifying low-frequency, non-conserved C-to-U RNA editing sites.

The input mitochondrial protein-coding gene sequences should be in FASTA format. Here is a sample: FASTA File