Skip to main content
Fig. 2 | Genome Biology

Fig. 2

From: MAAPER: model-based analysis of alternative polyadenylation using 3′ end-linked reads

Fig. 2

Accurate and sensitive PAS prediction by MAAPER. A Schematic of experimental design using QuantSeq FWD (QS-FWD) and QuantSeq REV (QS-REV) to examine APA in NIH3T3 cells treated with arsenic stress (AS) or undergoing recovery for 4 h (RC4) or 8 h (RC8) after AS. NT, non-treated cells. B Left, distribution of read-PAS distances based on genes with a single PAS. Right, as in the left plot, except that genes are divided into >400 nt (top) and ≤400 nt (bottom) groups. C Number of genes with a varying number of predicted PASs by MAAPER from the QS-FWD data. Bars denote the average gene number across all samples, and error bars denote SD. D Proportions of predicted PASs and QS-FWD reads based on their genomic positions. Proportions are averaged across all samples. E Precision rates of MAAPER’s PAS prediction for genes with a varying number of PASs. Bars denote the average precision across all samples. Error bars denote SD. F Proportions of four types of reads based on PAS prediction by MAAPER as well PAS identification by using QS-REV data. Mean and SD are based on all samples. G Nucleotide frequencies around three types of predicted PASs listed in F

Back to article page