Repository navigation
plotFingerprint results depend on BAM file order — input sorting needed #1410
Description
The output of plotFingerprint varies depending on the order of BAM files provided in the -b parameter, even when all other arguments remain identical. This affects both the plotted fingerprint and the raw/quality metrics outputs.
For example, the two following commands differ only in BAM and label order:
plotFingerprint -b ID1_filtered_shifted.bam ID2_filtered_shifted.bam ID3_filtered_shifted.bam \
-plot bams_fingerprint.pdf \
-p 1 \
-l 'ID1' 'ID2' 'ID3' \
--extendReads --skipZeros \
--outRawCounts plotFingerprint.raw.txt \
--outQualityMetrics plotFingerprint.qmetrics.txtplotFingerprint -b ID3_filtered_shifted.bam ID2_filtered_shifted.bam ID1_filtered_shifted.bam \
-plot bams_fingerprint.pdf \
-p 1 \
-l 'ID3' 'ID2' 'ID1' \
--extendReads --skipZeros \
--outRawCounts plotFingerprint.raw.txt \
--outQualityMetrics plotFingerprint.qmetrics.txtYet, the output values are different (example below):
ex :
< ID1 0.09378069129010372 0.495037541644535 0.3139128094725511 0.0 0.8441872981700753 0.500194148356678 0.5780619843832019
< ID2 0.06696695249101738 0.49499669220481174 0.4336652314316469 0.0 0.8618137782561894 0.5026794267798673 0.610113953814954
< ID3 0.1291594430331159 0.4932775653633279 0.3207750269106566 1.966908847312887e-189 0.7711248654467169 0.5012198748086091 0.48126701308648095
---
> ID3 0.12720877527444388 0.49059151283469915 0.347135955831608 4.577699669843273e-96 0.7701863354037267 0.4925363505619147 0.4677414185704291
> ID2 0.0664847262121746 0.4928753619064219 0.4357487922705314 1.7808421833197564e-168 0.8542443064182195 0.5087629908428597 0.6063354668238865
> ID1 0.09515023035629501 0.49307920612922806 0.31815044858523117 1.2376769385685896e-178 0.8329882677708764 0.4997551395064236 0.5681266081275786Version : deeptools 3.5.1
Activity
WardDeb commentedon Sep 27, 2026
Thanks for the report, the reason for this is the inclusion of --extendReads (which estimated the fragsize from first bam file only). This is changed now in #1483 and wil be part of the next release.