LYFE Sciences · Project HERA
Variant Interpretation · Classification Report
Generated: 2026-05-07
Case ID: NM_001001890.2_c.319_323del_20260507_144521
Framework: ACMG/AMP 2015
Variant classification summary

NM_001001890.2:c.319_323del

RUNX1  · NP_001001890.1:p.(Ala107GlnfsTer2)  · NM_001001890.2
GRCh37: chr21:36252957 GCCAGC>G  ·  GRCh38: chr21:34880660 GCCAGC>G
Gene: RUNX1 Transcript: NM_001001890.2
Final call
Pathogenic
PVS1 very strong PM2 supporting PM5 supporting PP5 supporting
All criteria require review: For research and educational purposes only.
Gene
RUNX1
Transcript
NM_001001890.2
Protein
NP_001001890.1:p.(Ala107GlnfsTer2)
gnomAD AF
ClinVar
OncoKB
Likely Oncogenic
Interpretation summary
Generated evidence synthesis
1
The RUNX1 c.319_323del (p.Ala107GlnfsTer2; p.A107Qfs*2) variant has been reported in ClinVar and classified as Pathogenic by the ClinGen Myeloid Malignancy Variant Curation Expert Panel.
2
This variant is absent from gnomAD v2.1 and gnomAD v4.1, which is below the RUNX1 PM2_Supporting threshold of <=0.00005 and far below the BS1 and BA1 population thresholds.
3
This 5-bp deletion causes an early frameshift with a premature stop codon, and the exact expert-panel ClinVar record states that the default RUNX1 transcript consequence p.Ala134fs is predicted to undergo nonsense-mediated decay, supporting PVS1; the same record also applied RUNX1-specific PM5_Supporting because the frameshift is downstream of c.98.
4
SpliceAI predicts no significant splice impact for this variant (max delta score 0.03), which is below the RUNX1 PP3 threshold of 0.38 and consistent with the <=0.20 splice caveat used for PM5_Supporting.
Final determination: Tavtigian et.al., 2020 - Bayesian adaptation of Richards et.al., 2015 v3.1.0 point-based framework yields a total score of 11, which maps to Pathogenic under the specified Tavtigian-style ranges.
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 Met This variant is a 5-bp frameshift deletion that causes an early premature stop codon (p.Ala107GlnfsTer2; default RUNX1 transcript consequence p.Ala134fs). RUNX1 loss of function is an established disease mechanism, and the exact ClinGen Myeloid Malignancy VCEP ClinVar record states that this frameshift is predicted to undergo nonsense-mediated decay, supporting PVS1 at very strong strength.
clinvar cspec pvs1_gene_context pvs1_variant_assessment
PS1 Not assessed No established alternate nucleotide change producing the same protein consequence was identified from the retrieved evidence, so PS1 was not assessed.
PS2 Not assessed No confirmed de novo occurrence with both maternity and paternity established was identified for this variant, so PS2 was not assessed.
PS3 Not assessed No variant-specific functional study demonstrating abnormal RUNX1 activity for this exact deletion was identified in the retrieved evidence, so PS3 was not assessed.
oncokb
PS4 Not assessed No exact proband count meeting RUNX1 phenotypic criteria was established from the retrieved evidence, so PS4 was not assessed.
clinvar
PM1 Not met Although this deletion affects the RUNX1 Runt homology region, the retrieved expert-panel ClinVar record for this exact variant applied PVS1, PM2_Supporting, and PM5_Supporting and did not establish PM1 for this frameshift. Available evidence therefore does not support applying PM1 in this pass.
clinvar cspec
PM2 Met This variant is absent from gnomAD v2.1 and gnomAD v4.1. The exact ClinGen Myeloid Malignancy VCEP ClinVar record states that it is completely absent from population databases with at least 20x coverage for RUNX1, meeting RUNX1 PM2_Supporting.
clinvar cspec gnomad_v2 gnomad_v4
PM3 N/A PM3 is not applicable in the RUNX1 Myeloid Malignancy VCEP specification.
cspec
PM4 N/A This variant is a frameshift deletion. In the RUNX1 VCEP specification, PM4 is used for in-frame insertions or deletions and stop-loss variants, so PM4 is not applicable here.
cspec
PM5 Met The exact ClinGen Myeloid Malignancy VCEP ClinVar record states that this frameshift is downstream of c.98 and applied PM5_Supporting under the RUNX1-specific rule for nonsense or frameshift variants in this region. SpliceAI predicts no significant splice impact (max delta score 0.03), which is below the <=0.20 splice caveat used in the RUNX1 PM5 framework.
clinvar cspec spliceai
PM6 Not assessed No assumed de novo occurrences without full parental confirmation were identified for this variant, so PM6 was not assessed.
PP1 Not assessed No segregation data with countable informative meioses were identified for this variant, so PP1 was not assessed.
PP2 N/A PP2 is not applicable in the RUNX1 Myeloid Malignancy VCEP specification.
cspec
PP3 N/A This variant is a frameshift deletion rather than a missense, synonymous, or intronic variant in the categories specified for RUNX1 PP3. SpliceAI shows no significant splice effect (max delta score 0.03), so PP3 is not applicable.
cspec spliceai
PP4 N/A PP4 is not applicable in the RUNX1 Myeloid Malignancy VCEP specification because the phenotype is not considered sufficiently specific for a single genetic etiology.
cspec
PP5 Met Expert panel ClinGen Myeloid Malignancy Variant Curation Expert Panel classified as Pathogenic.
cspec clinvar
BA1 Not met This variant is absent from gnomAD v2.1 and gnomAD v4.1 and therefore does not reach the RUNX1 BA1 threshold of >=0.0015 in any general population dataset.
cspec gnomad_v2 gnomad_v4
BS1 Not met This variant is absent from gnomAD v2.1 and gnomAD v4.1 and therefore does not fall within the RUNX1 BS1 range of 0.00015 to 0.0015 in a qualifying general population dataset.
cspec gnomad_v2 gnomad_v4
BS2 N/A BS2 is not applicable in the RUNX1 Myeloid Malignancy VCEP specification because RUNX1-related disease has incomplete penetrance and later onset of hematologic malignancy.
cspec
BS3 Not assessed No well-established study showing normal function for this exact variant was identified, so BS3 was not assessed.
oncokb
BS4 Not assessed No informative nonsegregation data were identified for this variant, so BS4 was not assessed.
BP1 N/A BP1 is not applicable in the RUNX1 Myeloid Malignancy VCEP specification.
cspec
BP2 Not assessed No evidence was identified showing this variant in trans with a pathogenic variant or in cis with a pathogenic variant, so BP2 was not assessed.
BP3 N/A BP3 is not applicable in the RUNX1 Myeloid Malignancy VCEP specification.
cspec
BP4 N/A This variant is a frameshift deletion rather than a missense, synonymous, or intronic variant in the categories specified for RUNX1 BP4. Although SpliceAI is low (max delta score 0.03), BP4 is not applicable for this variant type.
cspec spliceai
BP5 N/A BP5 is not applicable in the RUNX1 Myeloid Malignancy VCEP specification.
cspec
BP6 N/A BP6 is not applicable in the RUNX1 Myeloid Malignancy VCEP specification.
cspec
BP7 N/A This variant is a coding frameshift deletion, not a synonymous or intronic variant in the RUNX1 BP7 categories, so BP7 is not applicable.
cspec
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.