Identification of Long Non-coding RNA from inherent features using Machine Learning Techniques

Long non-coding RNAs are a distinctive class of non-coding RNAs of length greater than 200 nucleotides and no protein coding potential. LncRNA plays an important role in genetic and epigenetic regulation. Major studies reveal that IncRNAs are less conserved in their primary sequences and shows more functional characteristics at secondary structure level. The objective of this work is to identify an optimal sequencestructure combination for computational analysis of IncRNAs. We also propose a novel secondary structure quantization which consider the existence of various structure elements. The feature combinations when used as input to classification of IncRNAs from coding RNAs, significant improvement in the results were obtained.