Most of mRNAs and well-characterized long noncoding RNAs are shaped with 5' cap and 3' poly(A) tail. Thereby, conventional transcriptome analysis typically involved the enrichment of poly(A)+ RNAs by oligo(dT) selection. However, accumulated lines of evidence suggest that there are many RNA transcripts processed in alternative ways, which largely failed to be detected by oligo(dT) purification. Here, we describe an enrichment strategy to purify non-polyadenylated (poly(A)-/ribo-) RNAs from total RNAs by removal of poly(A)+ RNA transcripts and ribosomal RNAs. In the combination with high-throughput sequencing methods, this strategy has been successfully applied to identify the rich repertoire of non-polyadenylated RNAs in vivo.