FAQ
What genome build is the AutoPVS1 based on?
AutoPVS1 is now compatible with hg19/GRCh37 and hg38/GRCh38.
Does AutoPVS1 support users to select the biologically relevant transcript?
Transcript selection is crucial for all null variant types because variants can have different consequences for each transcript. Although we pre-selected transcript for each gene in AutoPVS1, it should be noted that multiple biological transcript isoforms may exist for one gene.
When users query a variant in a specific transcript (HGVS), AutoPVS1 regards this transcript as a biologically relevant transcript by default, and executes the algorithm based on the transcript provided by users.
Why does AutoPVS1 report back different results when users query the chromosomal position and HGVS of a same variant?
To fulfill the need that the users may choose the transcript by themselves, we set two different scenarios. When users query the chromosomal position, AutoPVS1 reports back the annotations and PVS1 strength level based on the transcript we pre-selected. On the other hand, if users query a variant in a specific transcript (HGVS), AutoPVS1 regards the transcript as a biological transcript by default, and executes the algorithm based on the transcript provided by users.
For example, 9-21971208-C-T corresponds to NM_058195.3:c.194-1G>A and NM_000077.4:c.151-1G>A in CDKN2A gene. When users query 9-21971208-C-T, autoPVS1 reports back NM_000077.4:c.151-1G>A because NM_000077.4 is the transcript we pre-selected. If users think NM_058195.3 is more biologically relevant and query NM_058195.3:c.194-1G>A, autoPVS1 will reported back PVS1 strength level based on transcript NM_058195.3.
How to search copy number variant (CNV)?
Currently, AutoPVS1 only support single-gene-level CNV as decribed in
Tayoun et al. (2018) [1].
To search a CNV in AutoPVS1, the CNVTYPE must be set to one of the currently recognised values (hg19):
| CNVTYPE | Description | Example |
|---|---|---|
| DEL | deletion | 11-2797090-2869333-DEL or del(11:2797090-2869333) |
| DUP | duplication is presumed in tandem | 1-215844214-215849058-DUP or dup(1:215844214-215849058) |
| TDUP | duplication is proven in tandem | 17-7577936-7585058-TDUP or tdup(17:7577936-7585058) |
| NTDUP | duplication is proven not in tandem | 17-7577936-7585058-NTDUP or ntdup(17:7577936-7585058) |
Why is the number of null variants in a gene less than expected in the web of AutoPVS1 when a gene name is queried?
Theoretically, the null variants of a gene are infinite and unpredicted. For example, a frameshift can occur in a chromosomal positive following 1 nucleotide deletion, 2 nucleotides deletion, and 3 nucleotides deletion, et al. Therefore, it is hard to list all the null variants. Currently, we only pre-annotate and list ~60,000 null variants retrieved from ClinVar (archive 2021-06). For variants not found in ClinVar, AutoPVS1 can still execute the program and assign a PVS1 strength level. These variants are not available in the variant list when you query a gene.
What is the decision path?
ClinGen’s Sequence Variant Interpretation Working Group proposed a decision tree to address variant-specific issues in the process of PVS1 interpretation [1]. Following the decision path, a preliminary PVS1 strength level was assigned. We label a code for each path. For example, if a nonsense variant is predicted to undergo nonsense-mediated decay (NMD) and preset in an exon in a biological transcript, the path is named as Nonsense/Framshift 1(NF1). IC1 and IC2 are not originated from the recommendation of the ClinGen’s SVI Working Group. More details can be found here [3].
What is pLI?
We adopted pLI from the Exome Aggregation Consortium (ExAC) [2]. The value presents the loss of function tolerance/intolerance of genes. The closer pLI is to one, the more LoF intolerant the gene appears to be. More details can be found at gnomad-v2-1
Which ClinVar version is used?
ClinVar archive 2021-06 is used in AutoPVS1.
Reference
[3] Jiale Xiang, Jiguang Peng, Zhiyu Peng. (2019). AutoPVS1: An automatic classification tool for PVS1 interpretation of null variants. Biorxiv.
[1] Abou Tayoun, A.N., Pesaran, T., DiStefano, M.T., Oza, A., Rehm, H.L., Biesecker, L.G., Harrison, S.M., and ClinGen Sequence Variant Interpretation Working Group. (2018). Recommendations for interpreting the loss of function PVS1 ACMG/AMP variant criterion. Hum Mutat 39, 1517-1524.
[2] Lek, M., Karczewski, K.J., Minikel, E.V., Samocha, K.E., Banks, E., Fennell, T., O'Donnell-Luria, A.H., Ware, J.S., Hill, A.J., Cummings, B.B., et al. (2016). Analysis of protein-coding genetic variation in 60,706 humans. Nature 536, 285-291.