Wolfram Data Repository

Immediate Computable Access to Curated Contributed Data

Alpha Missense

Source Notebook

Categorization of the pathogenicity data of human missense variants

Details

This data is developed by AlphaMissense model.
The default content is a Dataset containing a summary of the chromosomes information with the following properties for each chromosome:
"Chromosome"string including "Chromosome" and a number from 1 to 22 or a letter M,X,or Y.
"RawDataSize"total rows for each chromosome in the original database
"TotalPositions"total positions for each chromosome
"TotalMutations"total missense mutations for each chromosome
"Genome"the genome build
"TotalUniprotID"total uniprot IDs
"TotalTranscriptID"total transcript IDs
"TotalAminoAcidVariations"total amino acid changes
"PathogenicityQuartiles"pathogenicity Quartiles 1/4, 2/4, 3/4
"PathogenicityMean"pathogenicity Mean
"PathogenicityLikelyBenign"percentage of "likely_benign" classifications
"PathogenicityLikelyPathogenic"percentage of "likely_pathogenic" classifications
"PathogenicityAmbiguous"percentage of "ambiguous" classifications
Additional content elements are the original database for each chromosome:
"Chromosome"string including "Chromosome" and a number from 1 to 22 or a letter M,X,or Y.
"Position"genome position (1-based)
"ReferenceNucleotide"reference nucleotide (GRCh38.p13 for hg38)
"AlternativeNucleotide"alternative nucleotide
"Genome"genome build
"UniprotID"UniProtKB accession number of the protein in which the variant induces a single amino-acid substitution (UniProt release 2021_02)
"TranscriptID"Ensembl transcript ID from GENCODE V32 (hg38)
"AminoAcidVariation"Amino acid change induced by the alternative allele,in the format: Reference aminoacid-POS_aa-Alternative amino acid
"Pathogenicity"predicted probability of a variant being clinically pathogenic
"Classification"derived using the following thresholds: "likely_benign" for Pathogenicity < 0.34; "likely_pathogenic" for Pathogenicity > 0.564; and "ambiguous" otherwise
The EntityStore defined by proteins, named “AlphaMissense data organized by protein” has the following elements:
"ExternalIdentifier"ExternalIdentifier of the protein
"Name"common name of the protein
"Sequence"amino acid sequence of the protein
"Mutations"all possible mutations in the protein
"Score"AlphaMissense pathogenicity score by mutations of a protein
"Status"status of pathogenicity (B: Likely benign, A: Ambiguous, P: Likely pathogenic)
"MeanPathogenicity"Mean pathogenicity per residue of a protein
"MedianPathogenicity"Median pathogenicity per residue of a protein

Examples

Basic Examples (2) 

In[1]:=
ResourceData[\!\(\* TagBox["\"\<Alpha Missense\>\"", #& , BoxID -> "ResourceTag-Alpha Missense-Input", AutoDelete->True]\)][[;; 4]]
Out[1]=

Retrieve the data for ChromosomeX:

In[2]:=
chrX = ResourceData[\!\(\* TagBox["\"\<Alpha Missense\>\"", #& , BoxID -> "ResourceTag-Alpha Missense-Input", AutoDelete->True]\), "ChromosomeX"]; chrX[[;; 4]]
Out[3]=

Get a random sample of the Positions in ChromosomeX:

In[4]:=
RandomSample[Union@Normal@chrX[All, #Position &], 5]
Out[4]=

Get a random sample of the transcriptIDs in ChromosomeX:

In[5]:=
RandomSample[Union@Normal@chrX[All, #TranscriptID &], 5]
Out[5]=

Get a distribution of the ChromosomeX pathogenicity level for all possible missense variations:

In[6]:=
BoxWhiskerChart[chrX[All, #Pathogenicity &], "Mean", PlotLabel -> "Pathogenicy level distribution", FrameLabel -> {"ChromosomeX", "Pathogenicity level"}]
Out[6]=

Get a summary of the classification of the ChromosomeX pathogenicity level for all possible missense variations:

In[7]:=
PieChart[#[[All, 2]], ChartLegends -> #[[All, 1]]] &@ Tally[Normal@chrX[All, #Classification &]]
Out[7]=
Register the entity store:
In[8]:=
EntityRegister[ResourceData[\!\(\* TagBox["\"\<Alpha Missense\>\"", #& , BoxID -> "ResourceTag-Alpha Missense-Input", AutoDelete->True]\), "AlphaMissense data organized by protein"]]
Out[8]=

Select a protein from the set:

In[9]:=
protein = RandomEntity["AlphaMissenseAASubstitutionsProteins"]
Out[9]=

Find the protein's properties:

In[10]:=
protein["Properties"]
Out[10]=

Get the protein's name:

In[11]:=
protein["Name"]
Out[11]=

Get the protein's amino acid sequence:

In[12]:=
protein["Sequence"]
Out[12]=

Visualize the mean and median AlphaMissense pathogenicity per residue:

In[13]:=
ListLinePlot[{protein["MeanPathogenicity"], protein["MedianPathogenicity"]}, PlotLegends -> {"Mean", "Median"}, AxesLabel -> {"Residues", "Pathogenicity"}]
Out[13]=

Get the AlphaMissense pathogenicity score for a specific residue:

In[14]:=
resNumber = 65; AssociationThread[protein["Mutations"][[resNumber]], protein["Score"][[resNumber]]]
Out[15]=

Get the AlphaMissense pathogenicity status for a specific residue:

In[16]:=
pathAssoc = <|"B" -> "Likely benign", "A" -> "Ambiguous", "P" -> "Likely pathogenic"|>; AssociationThread[protein["Mutations"][[resNumber]], Map[pathAssoc, Characters[protein["Status"][[resNumber]]]]]
Out[17]=

Get the name of a specific protein:

In[18]:=
Entity["AlphaMissenseAASubstitutionsProteins", "X6R8D5"]["Name"]
Out[18]=

Scope & Additional Elements (2) 

Get the pathogenicity information associated to ChromosomeX in position 71765227:

In[19]:=
PathogenicityData[chromosome_String, mutationPosition_Integer, "Pathogenicity"] := Block[{rawData0}, rawData0 = ResourceData[\!\(\* TagBox["\"\<Alpha Missense\>\"", #& , BoxID -> "ResourceTag-Alpha Missense-Input", AutoDelete->True]\), chromosome][ Select[MatchQ[#[[2]], mutationPosition] &]]; Dataset[Association[{"Position" -> #[[2]], "Mutation" -> {#[[3]] -> #[[4]]}, "TranscriptID" -> #[[7]], "AminoAcidVariation" -> #[[8]], "Pathogenicity" -> #[[9]], "Classification" -> #[[10]]}] & /@ Normal[rawData0]]]
In[20]:=
PathogenicityData["ChromosomeX", 71765227, "Pathogenicity"]
Out[20]=

Get the pathogenicity information associated to ChromosomeX for a list of positions:

In[21]:=
PathogenicityData[chromosome_String, mutationPosition_List, "Pathogenicity"] := Block[{rawData0}, rawData0 = ResourceData[\!\(\* TagBox["\"\<Alpha Missense\>\"", #& , BoxID -> "ResourceTag-Alpha Missense-Input", AutoDelete->True]\), chromosome][ Select[MemberQ[mutationPosition, #[[2]]] &]]; Dataset[Association[{"Position" -> #[[2]], "Mutation" -> {#[[3]] -> #[[4]]}, "TranscriptID" -> #[[7]], "AminoAcidVariation" -> #[[8]], "Pathogenicity" -> #[[9]], "Classification" -> #[[10]]}] & /@ Normal[rawData0]]]
In[22]:=
PathogenicityData["ChromosomeX", {20193531, 20175194, 19372662, 31444512, 24179596}, "Pathogenicity"]
Out[22]=

Visualizations (1) 

Get the pathogenicity distribution associated to all the positions of all the chromosomes:

In[23]:=
data = Table[Normal[ResourceData[\!\(\* TagBox["\"\<Alpha Missense\>\"", #& , BoxID -> "ResourceTag-Alpha Missense-Input", AutoDelete->True]\), "Chromosome" <> ToString[i]][ All, #Pathogenicity &]], {i, {Sequence @@ Range[1, 22], "M", "X", "Y"}}];
In[24]:=
BoxWhiskerChart[data, "Mean", ChartLabels -> {Sequence @@ Table["chr" <> ToString[i], {i, 1, 22, 1}], "chrM", "chrX", "chrY"}, PlotLabel -> "Pathogenicy level distribution", FrameLabel -> {"", "Pathogenicity level"}]
Out[24]=

Wolfram Research, "Alpha Missense" from the Wolfram Data Repository (2024)  

Data Resource History

  • Date Created:

Source Metadata

  • Title: Science
  • Creator: Jun Cheng et al.
  • Date: 1 January 2023
  • Language: English
  • Citation:
    • Jun Cheng et al., Accurate proteome-wide missense variant effect prediction with AlphaMissense. Science 381, eadg7492(2023). DOI:10.1126/science.adg7492

Publisher Information

  • Prepared for the Wolfram Data Repository By: Lina Marcela Ruiz Galvis, Keiko Hirayama, Jason Biggs and Soutick Saha
  • Publisher of Record: Wolfram Research