Additional file 7 of Computational prediction of lncRNA-mRNA interactions by integrating tissue specificity in human transcriptome

Our predictions of TINCR-mRNA interactions using 31 different tissue-specific candidate mRNAs. For each tissue, the tissue-specific candidate mRNAs were selected by using RNA-seq data derived from Human Protein Atlas project (Expression Atlas ID: E-MTAB-2836). Combination of two prediction (ranking) methods (MinEnergy and SumEnergy) and two candidate mRNA sets (initial and tissue-specific) were used for the predictions. Experimentally-validated TINCR-mRNA interactions [9] (considered as true positives) were used for evaluating the prediction results. Horizontal axis indicates the number of predicted TINCR-mRNA interactions. Vertical axis indicates the total number of experimentally-validated interactions (true positives). The prediction using skin-specific candidates is already shown in Fig 2. (PDF 51 kb)