|Home | About | Journals | Submit | Contact Us | Français|
High throughput DNA sequencing is a powerful tool for profiling miRNA expression but presents many computational challenges. We have used xenograft osteosarcoma tumor lines with SOLiD sequencing technology to identify expression of miRNAs that is altered by anti-IGF1R antibody treatment. The quality of the sequencing runs was assessed with quality values for color space calls, total reads, correlation between biological replicates of mapped reads and percent mapped reads. To take advantage of the low error rate for base calling of SOLiD sequencing technology, sequence alignments must be performed in color space. In addition, the size range of miRNAs (18-28 bp) must be considered when aligning reads. The Small RNA Analysis Tool (RNA2MAP v 0.5), an Applied Biosystems free open source software tool that takes these points into account, was used to filter out reads resulting from adapters, transfer RNA (tRNA), ribosomal RNA, repetitive elements (including Alu, LINE and LTR), centromeric and telomeric satellites, small cytoplasmic RNA (scRNA) and small nuclear RNA (snRNA). Remaining reads were sequentially aligned to miRNAs present in miRBase v14 and then to the human genome reference sequence (assembly GRCh37) using RNA2MAP. Differentially expressed miRNAs, determined with a rank product non-parametric method using RankProd (a Bioconductor package), include hsa-mir-654 and hsa-mir-370. Interestingly, IRS1 mRNA is a TargetScan predicted target of hsa-mir-370 and IRS1 protein interacts with IGF1R. Throughout this analysis, several tools including custom scripts were needed to parse and format data. Differentially expressed miRNAs and their target genes may provide insight into the susceptibility of some osteosarcoma tumors and the resistance of others to anti-IGF1R antibody treatment. These studies may also lead to the identification of biomarkers for drug resistance.