Sample Total_reads Uniq_Rate Multi_Rate Unmap_Rate Uniq_Mapped_Reads Gene_Rate Ambiguity_Rate No_Feature_Rate Total_Tags CDS_Exon 5UTR_Exon 3UTR_Exon Introns Intergenic Counted_Reads Norm_Factor Avg_Corr Corr_diff MADScore isOutlier R1 43524087 78.26 13.22 8.52 34060708 80.25 4.75 15 42572577 56.41 6.8 24.09 11.1 1.61 27334231 1.0708 0.9646 0.005778 0.41416 N R2 42648752 78.42 13.65 7.93 33444783 82.44 4.81 12.75 42276165 57.9 6.47 24.64 9.63 1.37 27572067 0.9515 0.9616 0.002119 -0.04447 N R3 44514457 77.7 13.11 9.19 34587901 79.39 4.56 16.05 43211110 56.38 6.96 23.11 11.94 1.61 27460837 1.0697 0.9651 0.006318 0.48174 N R4 41609509 78.81 13.32 7.87 32792470 82.42 4.84 12.74 41352056 58.66 6.61 23.8 9.6 1.33 27027299 0.94 0.9622 0.002828 0.04447 N R5 34314210 77.26 13.3 9.44 26509694 81.76 4.81 13.42 33778915 58.92 6.69 23.21 9.7 1.48 21674911 1.1355 0.9543 -0.0066 -1.13709 N R6 41400550 79.67 12.62 7.71 32984639 83.17 4.97 11.86 41955210 60.86 6.2 22.89 8.83 1.21 27434527 1.0079 0.9676 0.009394 0.86724 N R7 39405548 81.86 12.02 6.12 32258277 78.2 4.35 17.46 39760217 55.81 6.71 22.87 12.9 1.71 25224557 1.0473 0.9556 -0.005024 -0.93955 N R8 46481525 82.5 12.57 4.93 38345133 82.91 4.74 12.35 48594850 60.5 6.33 22.77 9.14 1.26 31792234 0.8739 0.952 -0.009313 -1.47707 N R9 42081074 82.12 11.84 6.04 34556240 77.25 4.26 18.49 42341442 54.85 6.71 22.91 13.77 1.77 26694444 1.0427 0.9555 -0.005142 -0.95443 N R10 47368266 82.36 12.56 5.08 39010500 82.13 4.69 13.18 49300474 59.49 6.35 23.04 9.81 1.3 32037871 0.8799 0.9506 -0.011008 -1.6895 N R11 39710287 79.9 12.4 7.7 31726832 80.17 4.52 15.32 39724643 57.23 6.52 23.4 11.35 1.49 25434399 1.0635 0.9642 0.005234 0.34589 N R12 41676367 81.74 12.08 6.18 34067425 82.82 4.7 12.48 42989998 60.3 6.01 23.11 9.38 1.2 28215628 0.9551 0.9643 0.005416 0.36879 N

Controls: Brush - Drag vertically along an axis. Remove Brush - Tap the axis background. Reorder Axes - Drag an axis label horizontally. Color Lines by Values - Click on an axis label. Invert Axis - Double click on an axis label. Remove Axis - Drag axis label off the left or the right edge. Tooltip - Click on a line to show values, move out of the canvas to remove tooptip.

Total number of sequencing reads from raw fastq file

Percentage of reads uniquely mapped to the reference genome.

Percentage of reads mapped equally well to multiple locations on the reference genome.

Percentage of reads not mapped to the reference genome.

Number of uniquely mapped reads.

Percentage of reads mapped to unique gene region, including Exons, UTRs and Introns.

Percentage of reads mapped to gene-overlapping region and can't be unambiguously assigned to either gene.

Percentage of reads mapped to unannotated genomic regions.

Total number of tags (reads, one paired-end read counted as two reads)

Percentage of mapped reads on Exons.

Percentage of mapped reads on 5' UTR.

Percentage of mapped reads on 3' UTR.

Percentage of mapped reads on Introns.

Percentage of mapped reads on Intergenic regions.

Reads counted by featureCounts (=uniquely mapped read * uniq_rate)

Normalization factor calculated by edgeR

Average correlation where this sample is involved

Avg_Corr : Average correlation where this sample is involved;

Avg_Rest : Average correlation where this sample is NOT involved;

Corr_diff : The difference between Avg_Corr and Avg_Rest.

(1) For each sample, calculate the correlation difference. This is simply a difference between the average of all the pair wise correlations that involve the sample (for the same group) and the average of all the pair wise correlations that do not involve the sample. For example, if we have a, b, c, d for group 1, the correlation difference of sample a is:

The difference of Average (correlation(a, b), correlation(a, c), correlation(a, d)) and Average (correlation(b, c), correlation (b, d), correlation(c, d)). You can see that if sample a is an outlier, then the difference will be negative.

(2) Now we have a vector of values (one for each sample). We simply convert this vector to MAD scores (robust Z-scores) by subtracting the medians then dividing it by median absolute deviations (MAD). We use a standard MAD cutoff (e.g. -5) to determine the outliers.