search

OstfriesenBI / PredmiRNA

A set of scripts and tools to train a classifier for pre-miRNA Recognition
1 stars 0 forks source link

Feature calculation: Stacking triplets #15

Closed Finesim97 closed 5 years ago

Finesim97 commented 5 years ago

R/Python function: Input: csv file with the sequences and the secondary structure:

"comment","sequence","realmiRNA","secstructure",
"mmu-mir-380 MI0000797 Mus musculus miR-380 stem-loop","AAGAUG",1,"...(.."
"mmu-mir-381 MI0000798 Mus musculus miR-381 stem-loop","AAUUC",1,"...(.."

Source paper: Classification of real and pseudo microRNA precursors using local structure-sequence features and support vector machine See figure 1 and 2.

The best features seem to be the A(((, U(((, G((( and C((( triplets, but the calculation of all 32 variants would be nice.

mariusrueve commented 5 years ago

Paper

Figure1:

image

Figure2:

image