The link between bioinformatics approaches and experimental analysis for microrna target prediction in non-small cell lung cancer

Elham Tafsiri,1,* Mansour poorebrahim,2 Vahdat poortahmasebi,3 Parisa esmaeili kordlar,4 William cs cho,5



Lung cancer is the first cancer-related cause of death in the world. micrornas are small non-coding small molecules which have key roles in proliferation, apoptosis, invasion and metastasis, which could have more than one target gene at mrna level.


Some online tools like targetscan, mirdip, mirmap and miranda were used to evaluate the validated target genes. then before choosing target genes we took advantage of some bioinformatics tools including, string, genemania and tred to predict the target genes. finally, the expression level of target genes was measured in non- small-cell lung cancer (nsclc) tumor and adjacent normal tissues via sybr green qrt-pcr.


According to bioinformatics tools, bcl2 and akt3 were selected as target genes for mir-15/16; bcl-2 showed a significant negative correlation with mir-15a. furthermore, we found a significant correlation between bcl2 expression level and stage (p-value=0.04). pten was assumed as validated target gene of mir-21 that had a significant decrease in tumor tissues compared to adjacent normal tissues. irs1 was assigned as a target gene of mir-126/mir-128, at last hif1a was selected for mir-210. there was a significant negative association between irs1 expression level and mir-126, also a significant correlation was demonstrated between mir-210 and hif1a at mrna level.


Restoration of mir-15/16, mir-126 and mir-128 in nsclc might be therapeutic candidates to control cell proliferation and apoptosis.


Non-small-cell lung