LYFE Sciences · Project HERA
Variant Interpretation · Classification Report
Generated: 2026-08-02
Case ID: NM_006445.3_c.3299_14T_C_20260802_164311
Framework: ACMG/AMP 2015
Variant classification summary

PRPF8  · NP_006436.3:p.?  · NM_006445.3
GRCh37: chr17:1577722 A>G  ·  GRCh38: chr17:1674428 A>G
Gene: PRPF8 Transcript: NM_006445.3
Final call
All criteria require review: For research and educational purposes only.
Variant details
Gene
PRPF8
Transcript
NM_006445.3
Protein
NP_006436.3:p.?
gnomAD AF
0.0025099760228136786 (v4.1)
ClinVar
Benign
OncoKB
Classification rationale
Interpretation summary
Generated evidence synthesis
1
NM_006445.3:c.3299+14T>C is an intronic variant in PRPF8 located at position +14 of intron 21. The variant is present at extremely high frequency in population databases, with a grpmax filtering allele frequency of 4.48-4.65% in gnomAD and up to 4.78% in the African/African American subpopulation. This exceeds the BA1 stand-alone benign threshold of >1% by a wide margin.
2
The variant has been observed in 34 homozygous individuals in gnomAD v2.1 and 87 homozygous individuals in gnomAD v4.1, which is incompatible with autosomal dominant retinitis pigmentosa caused by PRPF8. This satisfies BS2 at strong benign strength.
3
SpliceAI predicts no splicing impact (max delta score = 0.01), with no predicted alterations to donor or acceptor sites, supporting a benign interpretation (BP4). No functional studies or clinical case reports were identified for this variant in the reviewed literature.
4
This variant has been classified as Benign by three clinical laboratories in ClinVar (VariationID 321893), though the review status is single-submitter and does not meet the 3-star expert panel threshold required for PP5/BP6 application under the governing framework.
5
Based on BA1 (stand-alone benign) alone, this variant is classified as Benign. BS1, BS2, and BP4 provide additional independent evidence supporting benign classification.
Final determination:
Criteria assessment
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 N/A Intronic variant at position c.3299+14 in intron 21, not affecting the canonical splice acceptor or donor sites (±1,2). Does not fall into any PVS1 null-variant bucket (nonsense, frameshift, or canonical splice consensus). SpliceAI predicts no splicing impact (max delta = 0.01).
spliceai pvs1_variant_assessment
PS1 N/A Not applicable to intronic variants. PS1 requires a different nucleotide change at the same position with the same predicted amino acid effect, which cannot be evaluated for an intronic substitution.
PS2 Not met No de novo observation was identified in any reviewed publication or case data for NM_006445.3:c.3299+14T>C.
PS3 Not met No functional studies were identified for NM_006445.3:c.3299+14T>C. None of the reviewed publications contain variant-specific or systematically characterized range-based functional data for this intronic variant. SpliceAI predicts no splicing impact (max delta = 0.01), but this is in silico evidence (BP4), not experimental functional data satisfying PS3.
spliceai
PS4 Not met No case-control studies or systematic cohort evidence was identified for this variant. The variant is too common in population databases (gnomAD v2.1 AF = 0.455% overall, 4.74% in African/African American) to be enriched in a rare disease cohort. None of the reviewed publications report patient-level observations of this variant.
gnomad_v2 gnomad_v4
PS5 Not met ClinVar classification is Benign, not Pathogenic. No reputable source reports this variant as pathogenic. The two criterion-lead submissions (SCV001733155, SCV005251965) both classify as Benign.
clinvar
PM1 Not met This is an intronic variant at position c.3299+14, not located in a characterized functional domain or mutational hotspot. The variant does not alter any protein coding sequence. No hotspot significance at cancerhotspots.org for this residue.
PM2 Not met This variant is present in gnomAD at frequencies far exceeding the PM2 threshold of <0.1%. gnomAD v2.1 overall AF = 0.455% (1286/282376 alleles, 34 homozygotes); gnomAD v4.1 overall AF = 0.251% (4047/1612366 alleles, 87 homozygotes). The variant is a common polymorphism.
gnomad_v2 gnomad_v4
PM5 N/A Not applicable to intronic variants. PM5 requires a different missense change at the same residue with a pathogenic classification. This intronic variant has no amino acid residue context for comparator analysis.
pm5_candidates
PM6 Not met No de novo observation was identified in any reviewed publication or case data for NM_006445.3:c.3299+14T>C. None of the curated ClinVar submissions report de novo origin.
clinvar
PP1 Not met No segregation data was identified for this variant in any reviewed publication or case submission.
PP2 Not met This is an intronic variant, not a missense change. PP2 applies specifically to missense variants in genes with a low rate of benign missense variation where missense variants are a common disease mechanism. Not applicable to intronic substitutions.
PP3 Not met No in silico evidence supports a deleterious effect. SpliceAI predicts no splicing impact (max delta = 0.01). REVEL and BayesDel scores are not available for intronic variants. No computational tool predicts pathogenicity for this variant.
spliceai
PP4 Not met No specific patient phenotype data was provided for this case. No evidence that this variant was observed in an individual with a clinical presentation consistent with PRPF8-related disease (retinitis pigmentosa).
PP5 Not met ClinVar classification is Benign, not Pathogenic/Likely Pathogenic. PP5 requires a reputable source (3-star expert panel) reporting the variant as pathogenic. This variant is classified as Benign by three clinical laboratories with review status 'criteria provided, single submitter' — not an expert panel review. Per the governing framework, PP5 is only applied at supporting strength with ClinVar 3-star EP classification; this entry does not qualify.
clinvar
BA1 Met This variant has an allele frequency far exceeding the BA1 threshold of >1%. The grpmax filtering allele frequency is 4.48% in gnomAD v2.1 (African/African American subpopulation AF = 4.74%) and 4.65% in gnomAD v4.1 (African/African American subpopulation AF = 4.78%). Additionally, 34 homozygotes are observed in gnomAD v2.1 and 87 homozygotes in gnomAD v4.1, confirming this is a common polymorphism incompatible with a highly penetrant Mendelian disorder such as PRPF8-related retinitis pigmentosa.
gnomad_v2 gnomad_v4
BS1 Met The overall allele frequency exceeds the BS1 threshold of >0.3%. gnomAD v2.1 overall AF = 0.455% (1286/282376 alleles). While gnomAD v4.1 overall AF = 0.251% is below the BS1 threshold, the v2.1 frequency meets the criterion. This criterion is effectively subsumed by BA1 (stand-alone benign) given the grpmax FAF of 4.5%, but is independently met based on population frequency.
gnomad_v2
BS2 Met This variant has been observed in 34 homozygous individuals in gnomAD v2.1 and 87 homozygous individuals in gnomAD v4.1. PRPF8-related retinitis pigmentosa is inherited in an autosomal dominant manner. Observation of multiple healthy homozygous individuals for a dominant disorder is inconsistent with pathogenicity and satisfies BS2.
gnomad_v2 gnomad_v4
BS3 Not met No functional studies were identified demonstrating that NM_006445.3:c.3299+14T>C has no deleterious effect. While SpliceAI predicts no splicing impact (delta = 0.01), this is in silico computational prediction (BP4), not experimental functional evidence. No in vitro or in vivo functional assays have tested this variant or a systematically characterized range that includes this position.
spliceai
BS4 Not met No segregation data demonstrating lack of cosegregation with disease was identified for this variant. No family studies including this variant were found in the reviewed literature.
BP1 Not met This is an intronic variant, not a missense change. BP1 applies specifically to missense variants in genes where primarily truncating variants cause disease. Additionally, PRPF8 is associated with disease through both missense and truncating pathogenic variants, so BP1 would not apply even for a missense variant.
BP2 Not met No evidence was identified of this variant being observed in trans with a known pathogenic variant for a fully penetrant dominant disorder, or in cis with a pathogenic variant in any reviewed publication or case submission.
BP4 Met Multiple lines of computational evidence support a benign interpretation. SpliceAI predicts no splicing impact for this intronic variant (max delta score = 0.01), with no predicted acceptor gain, acceptor loss, donor gain, or donor loss. REVEL and BayesDel scores are not applicable to intronic variants. The SpliceAI prediction is specifically designed to detect splice-altering variants and finds no evidence of altered splicing at this position.
spliceai
BP5 Not met BP5 requires observation of the variant in a case with an alternate molecular basis for disease. No such case data was identified for this variant.
BP6 Not met ClinVar classification is Benign with review status 'criteria provided, single submitter' — not a 3-star expert panel review. Per the governing framework, BP6 requires ClinVar 3-star expert panel classification to apply at supporting strength. The ClinVar entry for this variant (VariationID 321893) does not meet the expert panel threshold.
clinvar
BP7 N/A BP7 is specifically defined for synonymous (silent) coding variants with no predicted splicing impact and low nucleotide conservation. This is an intronic variant at position c.3299+14, not a synonymous coding change. BP7 does not apply to intronic substitutions.
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.