UniParental Disomy (UPD) is genetic phenomenon where an individual inherits two copies of a chromosome from a single parent and none from the other. UPD can take the form of heterodisomy where the child inherits non identical chromosomes from one parent, or isodisomy where the child inherits a duplicate chromosome from a parent [1].
UPDOG is a Python tool to find and plot UPD events in Next Generation Sequencing (NGS) Whole Genome Sequencing trio data. Given a VCF containing a sample and the sample's two parents the tool will produce and plot a set of metics for each chromosome. The user can then examine these plots for signs of UPD events. Alternatively, the tool produces a CSV file containing the statistically significant UPD events.
- Examine each chromosome in segments e.g. 1MB and calculate the proportion of mendellian errors in each segment.
- Calculate the expected number of mendellian errors (Mean over all chromosomes).
- For each segment in step 1 measure whether each segment has a statistically significant increase in each type of mendelian error.
- Merge contiguous statistically significant segments together.
- Apply hard filters to the UPD calls and report to user.
Each UPD event produces signature mendellian errors in trio genotypes [2]
Coming Soon
Coming Soon
pip install pyupdog
UPDog.py --vcf test_data/200518_A00748_0027_AHLKFCDRXX.vcf.gz \
--proband_id test_sample \
--ped test_data/200518_A00748_0027_AHLKFCDRXX.ped \
--output results/200518_A00748_0027_AHLKFCDRXX \
--min_dp 25 \
--min_gq 15 \
--min_qual 20 \
--p_value 0.001 \
--block_size 1000000 \
--min_variants_per_block 100 \
--min_blocks 2 \
--min_proportion 0.01
Coming Soon
[1] https://en.wikipedia.org/wiki/Uniparental_disomy [2] UPDio https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3975066/