Note: This pipeline is under development, and the results may be wrong or inaccurate.
Accuracy is measured as the F1 score and shown for different read lengths and single/paired end reads. Further divided into reads containing and not containing genomic variants. A read contains a variant is it has been simulated so that it contains a variant allele not in the reference genome.
Recall and 1-Precision is shown for different subsets of the reads, depending on the MAPQ-score (shown as dots in the plot). For every MAPQ-score, only reads with that score or higher are included.
Showing read mapping accuracy for reads simulated with different error profiles (here described as low, medium and high error rate).
Peaks are simulated using Chips. The accuracy is the number of peaks found after calling peaks by MACS and selecting the top N peaks where N is the same number of peaks as in the simulated data.
Shown for single-end and paired-end reads and different read lengths.
Memory usage when mapping 2 million reads using 4 threads to hg38.
