Evaluation of microRNA alignment techniques

Abstract

Genomic alignment of small RNA (smRNA) sequences such as microRNAs poses considerable challenges due to their short length (∼21 nucleotides [nt]) as well as the large size and complexity of plant and animal genomes. While several tools have been developed for high-throughput mapping of longer mRNA-seq reads (>30 nt), there are few that are specifically designed for mapping of smRNA reads including microRNAs. The accuracy of these mappers has not been systematically determined in the case of smRNA-seq. In addition, it is unknown whether these aligners accurately map smRNA reads containing sequence errors and polymorphisms. By using simulated read sets, we determine the alignment sensitivity and accuracy of 16 short-read mappers and quantify their robustness to mismatches, indels, and nontemplated nucleotide additions. These were explored in the context of a plant genome (Oryza sativa, ∼500 Mbp) and a mammalian genome (Homo sapiens, ∼3.1 Gbp). Analysis of simulated and real smRNA-seq data demonstrates that mapper selection impacts differential expression results and interpretation. These results will inform on best practice for smRNA mapping and enable more accurate smRNA detection and quantification of expression and RNA editing.

MiR&moRe2: A Bioinformatics Tool to Characterize microRNAs and microRNA-Offset RNAs from Small RNA-Seq Data

“…Regarding miR&moRe2 pipeline design and implementation features, a series of filters on raw data are applied by means of efficient methods [40], and optimal parameters for read alignment are employed [41]. Moreover, miR&moRe2 makes use of the best performing and widely used methods for miRNA prediction, miRDeep2 [25] and RNAfold [42].…”

Accuracy, speed and error tolerance of short DNA sequence aligners

“…The accuracy of mappers with each test was quantified using the F-measure that considers both precision and accuracy. Throughout this paper, we use a beta value of 0.25 that weights precision higher than recall, as described previously [25](Ziemann et al, 2016) .…”

OPTIMIR, a novel algorithm for integrating available genome-wide genotype data into miRNA sequence alignment analysis

“…Read alignment process -For read alignment, we opted for the bowtie2 software (Langmead and Salzberg, 2012) that can handle trimming and tailing events at the reads' extremities via its local alignment mode which has been shown to be efficient for miRSeq data alignment (Ziemann et al, 2016). Only reads with a sequence of at least 17 consecutive nucleotides (defined as the alignment seed) that perfectly match with the reference library are kept in the analysis (see Supplementary Table 1 for details concerning the choice of the seed value and its consequences on isomiR detection).…”