LYFE Sciences · Project HERA
Variant Interpretation · Classification Report
Generated: 2026-05-01
Case ID: NM_022552.4_c.856-59_1014_13dup_20260501_135411
Framework: ACMG/AMP 2015
Variant classification summary

NM_022552.4:c.856-59_1014+13dup

DNMT3A  · NP_072046.2:p.?  · NM_022552.4
GRCh37: chr2:25470446 A>ACCCCACAACTTACCACTGAGAATTTGCCGTCTCCGAACCACATGACCCAGCGGGTGCCTTCAGCTGCTCGGCTCCGGCCCGTCATCCACCAAGACACAATGCGGCCTGGCCACCAGGAGAAGCCCCGCAGTTTCCCCCACACCAGCTCCCCAATGCCAAAGCCCCGGCCGTCCTGGAGCCCCAAGGAGCAGAAATCATTACACAGTGGTCACGAGGCCCTGCCACCCTGAT  ·  GRCh38: chr2:25247577 A>ACCCCACAACTTACCACTGAGAATTTGCCGTCTCCGAACCACATGACCCAGCGGGTGCCTTCAGCTGCTCGGCTCCGGCCCGTCATCCACCAAGACACAATGCGGCCTGGCCACCAGGAGAAGCCCCGCAGTTTCCCCCACACCAGCTCCCCAATGCCAAAGCCCCGGCCGTCCTGGAGCCCCAAGGAGCAGAAATCATTACACAGTGGTCACGAGGCCCTGCCACCCTGAT
Gene: DNMT3A Transcript: NM_022552.4
Final call
VUS
PM2 supporting PP3 supporting
All criteria require review: For research and educational purposes only.
Gene
DNMT3A
Transcript
NM_022552.4
Protein
NP_072046.2:p.?
gnomAD AF
ClinVar
OncoKB
Interpretation summary
Generated evidence synthesis
1
The DNMT3A c.856-59_1014+13dup (p.?) variant has not been observed in somatic cancers in COSMIC and has not been reported in ClinVar.
2
This variant is absent from gnomAD v2.1 and gnomAD v4.1, so its population frequency is below the non-VCEP PM2 threshold of 0.1%.
3
SpliceAI predicts a strong splice effect with a maximum delta score of 0.91, supporting possible abnormal splicing, although the exact transcript consequence of this intragenic duplication remains uncertain.
Final determination: Generic ACMG/AMP 2015 fallback rules do not meet benign, likely benign, likely pathogenic, or pathogenic combination thresholds, so the variant is classified as Variant of Uncertain Significance.
ACMG/AMP criteria review
Criteria shown when status is available
All criteria require review: For research and educational purposes only.
Criterion Status Rationale Evidence used
PVS1 Not met Loss of function is an established DNMT3A disease mechanism, but this duplication does not fall into the default generic PVS1 null-variant categories of nonsense, frameshift, or canonical +/-1,2 splice variants. The exact transcript consequence remains uncertain for this intragenic duplication, so PVS1 is not applied.
pvs1_gene_context pvs1_variant_assessment pvs1_generic_framework
PS1 N/A PS1 applies to a variant that results in the same amino acid change as an established pathogenic variant. This duplication has an uncertain protein consequence (p.?), so PS1 is not applicable.
PS2 Not assessed No confirmed de novo occurrence with verified maternity and paternity was identified for this variant, so PS2 cannot be assessed from the available evidence.
PS3 Not assessed No well-established functional study was identified for this variant. In particular, no RNA study or other assay was found to show whether this duplication causes abnormal splicing or another damaging effect.
PS4 Not assessed No evidence was identified showing that this exact variant is enriched in affected individuals compared with controls, so PS4 is not met.
clinvar
PM1 Not assessed No evidence was identified that this duplication affects a well-established critical functional domain or mutational hot spot without benign variation.
PM2 Met This variant is absent from gnomAD v2.1 and gnomAD v4.1, so the observed population frequency is below the non-VCEP PM2 threshold of 0.1%, supporting rarity.
gnomad_v2 gnomad_v4
PM3 N/A PM3 is used for recessive disorders with detected variants in trans. The available evidence does not support use of a recessive in-trans framework for this variant.
PM4 Not assessed This intragenic duplication spans intronic sequence on both sides of a coding segment, and the actual transcript and protein consequences remain uncertain. Because it is not established that the mature transcript contains a simple in-frame protein length change, PM4 is not applied.
pvs1_variant_assessment spliceai
PM5 N/A PM5 applies to a novel missense change at an amino acid residue where a different pathogenic missense change has been established. This duplication does not have a defined missense consequence, so PM5 is not applicable.
PM6 Not assessed No report was identified showing this variant to be apparently de novo without confirmed parentage, so PM6 cannot be assessed.
PP1 Not assessed No segregation data were identified for this variant, so PP1 cannot be assessed.
PP2 N/A PP2 is a missense-specific criterion. This duplication does not have an established missense consequence, so PP2 is not applicable.
PP3 Met SpliceAI predicts a strong splice effect for this variant, with a maximum delta score of 0.91, which is above commonly used concern thresholds and supports an abnormal splicing effect. REVEL and BayesDel were not available because this is not an SNV.
spliceai
PP4 Not assessed No phenotype or family history information was provided to determine whether the clinical presentation is highly specific for a DNMT3A-related disorder, so PP4 cannot be assessed.
PP5 Not assessed No reputable source classification for this exact variant was identified. This variant is absent from ClinVar, so PP5 is not met.
clinvar
BA1 Not met This variant is absent from gnomAD v2.1 and gnomAD v4.1 and therefore does not meet the stand-alone benign frequency threshold of 1%.
gnomad_v2 gnomad_v4
BS1 Not met This variant is absent from gnomAD v2.1 and gnomAD v4.1 and therefore does not meet the benign strong frequency threshold of 0.3%.
gnomad_v2 gnomad_v4
BS2 Not assessed No evidence was identified showing this variant in healthy adult individuals in a context sufficient to support BS2.
BS3 Not assessed No well-established functional study showing no damaging effect was identified for this variant, so BS3 cannot be applied.
BS4 Not assessed No non-segregation data were identified for this variant, so BS4 cannot be assessed.
BP1 N/A BP1 is a missense-specific criterion and is not applicable to this duplication.
BP2 Not assessed No phase data or additional variant data were identified to support BP2.
BP3 Not assessed No evidence was identified showing that this duplication lies within a repetitive region without known function in a way that would support BP3.
BP4 Not met Available computational evidence does not support a benign interpretation. SpliceAI predicts a substantial splice effect with a maximum delta score of 0.91, which argues against BP4.
spliceai
BP5 Not assessed No alternate molecular explanation was identified that would allow BP5 to be assessed.
BP6 Not assessed No reputable source benign classification was identified for this exact variant. This variant is absent from ClinVar, so BP6 is not met.
clinvar
BP7 Not met BP7 is not supported because computational evidence predicts splice disruption rather than no impact on splicing. SpliceAI shows a maximum delta score of 0.91.
spliceai
Disclaimer: The content and results provided by LYFE Sciences are for research and educational purposes only and must not be used as a substitute for professional medical judgment, diagnosis, or treatment. Always consult a qualified healthcare professional before making any clinical decisions.