I am essentially looking for gene to protein mapping/lookup table. I have gene names (in biomart, it's called external_gene_name to be specific). Looks something like this:
"PLPPR3" "TSPY10" "ELANE" "DENND11" "TSPY4" "CCL15" "PSMB3" "TAS2R46" "ZBTB9" "KIR2DL5A" "RWDD2B"
I want to know the proteins produced by these genes. The protein IDs should look like this (cell surface markers for example):
"CD45RA" "CD27" "CD16" "GPR56" "CD56" "CD57" "CD94" "CD158"
I am not sure what is the official terminology for these IDs. uniprot? swissprot? something else? Does anyone know where to find mappings for these gene names to protein ids? If they are on biomart, perhaps someone knows the name of the field? Thanks!
Update: This is the R code that I use to fetch the data:
library(biomaRt)
mart <- useMart("ENSEMBL_MART_ENSEMBL")
mart <- useDataset(mart=mart,dataset="hsapiens_gene_ensembl")
pdata <- getBM(mart=mart,attributes=c("external_gene_name","protein_id"),filters=c("biotype"),values=list("protein_coding"),useCache=FALSE)
external_gene_name protein_id
1
2 ABK41909
3 BAF82881
4 BAG36999
5 EAX09448
6 TMPRSS15 AAC50138
7 TMPRSS15 CAB65555
8 TMPRSS15 CAB90389
9 TMPRSS15 CAB90392
10 TMPRSS15 AAI11750
11 TMPRSS15
12 SMIM34B
13 GATD3B
14 GATD3B BAA20888
...
The attribute protein_id returns some strange looking protein IDs. It should be replaced with something else. Not sure what to use there.
2 answers
Gene names Entry name
1 CD63 MLA1 TSPAN30 CD63_HUMAN
2 CDV3 H41 CDV3_HUMAN
3 CD79A IGA MB1 CD79A_HUMAN
4 CD8B CD8B1 CD8B_HUMAN
5 CDKL2 CDKL2_HUMAN
6 CDRT4 CDRT4_HUMAN
7 CDX2 CDX3 CDX2_HUMAN
8 CDK11A CDC2L2 CDC2L3 PITSLREB CD11A_HUMAN
9 CD14 CD14_HUMAN
10 CD22 SIGLEC2 CD22_HUMAN
# … with 167 more rows
Mostly protein and gene names are the same. I have found some of the examples on your list under gene synonyms. The gene synonyms are basically any names that the gene/protein have been ever known as in the literature, they are not names from any specific database. The CD names you have are terms people use in the literature, they are not any official name used in any database, which makes them gene synonyms.
Wow! This is shocking! So the most commonly used "protein IDs" (cell surface markers, CDXXX) are just in people's minds and not official! If there is no database, I am not sure how to resolve two of my issues:
- All gene names are not same as protein names. Some are completely different. Few examples:
```
marker gene
Cd45 PTPRC
Cd11b ITGAM
Cd115 CSF1R
```
Update: As you said, they are old synonyms. I also found them on UniProt. For example:
- There must be a one-to-many relationship. One gene produces many proteins, so a gene name must match to multiple protein names. Or are all proteins from a gene given the same name?
Update: The closest I got to is to download the table from UniProt SwissProt Human.
Gene names Entry name
1 CD63 MLA1 TSPAN30 CD63_HUMAN
2 CDV3 H41 CDV3_HUMAN
3 CD79A IGA MB1 CD79A_HUMAN
4 CD8B CD8B1 CD8B_HUMAN
5 CDKL2 CDKL2_HUMAN
6 CDRT4 CDRT4_HUMAN
7 CDX2 CDX3 CDX2_HUMAN
8 CDK11A CDC2L2 CDC2L3 PITSLREB CD11A_HUMAN
9 CD14 CD14_HUMAN
10 CD22 SIGLEC2 CD22_HUMAN
# … with 167 more rows
Log in to answer this question.
That's an INSDC protein ID. See http://www.ensembl.org/Homo_sapiens/Transcript/Similarity?db=core;g=ENSG00000277117;r=21:5022493-5040666;t=ENST00000623960