Return to database

Back to top

Curatopes Melanoma documentation

Where can I find the peer-reviewed article on Curatopes Melanoma?

A Resource Report on Curatopes Melanoma was published in Cancer Research:

Lischer C, Eberhardt M, Jaitly T, Schinzel C, Schaft N, Dörrie J, Schuler G, and Vera J. Curatopes Melanoma: a database of predicted T-cell epitopes from overly expressed proteins in metastatic cutaneous melanoma. Cancer Res. 2019 Aug 15. [Epub ahead of print] PMID: 31416842. doi: 10.1158/0008-5472.CAN-19-0296.

How were the listed genes in the different tables selected?

The four available tables are Superior tolerance (predicted), Enhanced tolerance (predicted), All predicted epitopes, and Known melanoma epitopes. Superior tolerance (predicted) and Enhanced tolerance (predicted) are subsets of All predicted epitopes.

We based our analysis on a set of 32 metastatic cutaneous melanoma biopsies (all samples but GSM2069836 from GEO dataset GSE78220, and samples GSM2536673, -675, -677, -679, and -681 from GEO dataset GSE96619) and on tissue profiles from the Human Protein Atlas and the GTEx transcriptome set. All epitopes are annotated with relevant features and a score was calculated to rank epitopes according to their expected clinical performance.

How were epitopes predicted for the selected genes?

Annotated human protein sequences were obtained from Ensembl (release 94). Epitopes were predicted from these sequences with the offline-version of IEDB tools leveraging the included Python script using artificial neural networks and netMHCpan version 4.0.

Which features are listed in the tables?

Physicochemical features for each peptide include its amino acid sequence and length. In addition, listed features the official gene symbol and Ensembl transcript accession of the corresponding protein.

How can I find epitopes based on my own criteria?

Below the table headers, you will see empty boxes. Each of these boxes can be used to filter the corresponding column based on values that you specify, and multiple columns can be filtered against simultaneously.

How can I obtain bulk epitope data for my own analyses?

Below each table, you will find two buttons: Download Filtered Rows and Download All Rows. The latter will always download the complete table that you are currently inspecting while the former will download the collection of epitopes from that table that you have set up via filtering.

How is the gPIE (generalized-epitope Predicted Immuno-Efficacy) score calculated?

Among the predicted epitope features are binding affinity (between peptide and the HLA molecule) and immunogenicity (i.e., T-cell receptor engagement). These two features were combined with the observed median transcript expression and the gene expression index in melanoma biopsies to calculate the gPIE score.

In an intial step, binding affinity, immunogenicity, and median transcript expression were normalized to the value range between 0 and 1. The product of the normalized values was then multiplied with the gene expression index and with 100 to make the values more readable. See the formulas below for the exact mathematical treatment. $$ gPIE_{Epitope} = 100 \cdot f_1(BindAff_{Epitope}) \cdot f_2(Immunog_{Epitope}) \cdot f_3(MelTranscExp_{Peptide}) \cdot f_4(GeneExpIndex_{Peptide})$$ $$ f_1(BindAff_{Epitope}) = \frac{-BindAff_{Epitope} - min(-BindAff)}{max(-BindAff) - min(-BindAff)} = \frac{BindAff_{Epitope} - max(BindAff)}{min(BindAff) - max(BindAff)} $$ $$ f_2(Immunog_{Epitope}) = \frac{Immunog_{Epitope} - min(Immunog)}{max(Immunog) - min(Immunog)} $$ $$ f_3(MelTranscExp_{Peptide}) = \begin{cases} \frac{MelTranscExp_{Peptide}}{100} &\quad \text{if MelTranscExp}_{Epitope} \lt 100 \\ 1 &\quad\text{otherwise} \\ \end{cases} $$ $$ f_4(GeneExpIndex_{Peptide}) = \frac{MelGeneExp_{Peptide}} {MelGeneExp_{Peptide} + MaxTissueGeneExp_{Peptide}} $$ The gene expression values in f4 are the ones listed in the two table colums Lower-bound Tumor Gene Expression and Max Upper-bound Tissue Gene Expression. The achievable minimum and maximum of the gPIE score are 0 and 100, respectively.

Which genes made it into the tables?

Peptides from these genes are found in the corresponding tables:

Gene Symbol Gene Name Superior tolerance (predicted) Enhanced tolerance (predicted) All predicted epitopes Known melanoma epitopes
ATOX1 antioxidant 1 copper chaperone yesyes
C11orf71 chromosome 11 open reading frame 71 yesyes
CDKN2A cyclin-dependent kinase inhibitor 2A yes
CLPP caseinolytic mitochondrial matrix peptidase proteolytic subunit yes
COX7A2L cytochrome c oxidase subunit VIIa polypeptide 2 like yesyes
CSNK1A1 casein kinase 1, alpha 1 yes
CSNK2A3 casein kinase 2, alpha 3 polypeptide yesyes
CTAG1B cancer/testis antigen 1B yes
DCT dopachrome tautomerase yes
ELOC elongin C yesyes
FOLH1 folate hydrolase (prostate-specific membrane antigen) 1 yes
GAS7 growth-arrest-specific 7 yes
GPR143 G-protein-coupled receptor 143 yes
GPR89A G-protein-coupled receptor 89A yesyes
GPR89B G-protein-coupled receptor 89B yesyes
GTF2H2C GTF2H2 family member C yesyes
HAUS3 HAUS augmin-like complex, subunit 3 yes
HSBP1 heat-shock-factor-binding protein 1 yesyes
KRTCAP2 keratinocyte-associated protein 2 yesyes
MAGEA3 melanoma antigen family A3 yesyes
MCTS1 malignant T-cell-amplified sequence 1 yesyes
MLANA melan-A yes
MPLKIP M-phase-specific PLK1 interacting protein yesyes
MRPS33 mitochondrial ribosomal protein S33 yesyes
NACA nascent polypeptide-associated complex alpha subunit yesyes
NBPF20 neuroblastoma breakpoint family, member 20 yesyes
NDUFA11 NADH dehydrogenase (ubiquinone) 1 alpha subcomplex, 11, 14.7kDa yesyes
PCLAF PCNA-associated factor yesyes
PGAM4 phosphoglycerate mutase family member 4 yesyes
PIGF phosphatidylinositol glycan anchor biosynthesis, class F yesyes
PMEL premelanosome protein yes
PPP1R3B protein phosphatase 1, regulatory subunit 3B yes
PRAME preferentially expressed antigen in melanoma yes
PSMG4 proteasome (prosome, macropain) assembly chaperone 4 yesyes
RAB38 RAB38, member RAS oncogene family yes
RAMACL RNA guanine-7 methyltransferase-activating subunit-like (pseudogene) yesyes
RAN RAN, member RAS oncogene family yesyes
RIOX1 ribosomal oxygenase 1 yesyes
RPL37 ribosomal protein L37 yesyes
RPS19 ribosomal protein S19 yesyes
RPS27L ribosomal protein S27-like yesyes
SERF2 small EDRK-rich factor 2 yesyes
SMIM10L1 small integral membrane protein 10 like 1 yesyes
SMUG1 single-strand-selective monofunctional uracil-DNA glycosylase 1 yesyes
TIMM23B translocase of inner mitochondrial membrane 23 homolog B (yeast) yesyes
TMBIM4 transmembrane BAX inhibitor motif-containing 4 yesyes
TRAPPC9 trafficking protein particle complex 9 yesyes
TXNL4A thioredoxin-like 4A yesyes
TYR tyrosinase yes
USP41 ubiquitin-specific peptidase 41 yesyes

Which HLA alleles were surveyed?

We based our analysis on a list of 36 well-characterized HLA class I alleles that are common in the European population.

What are survival-critical tissues?

Our tissue analysis was based on the GTEx transcriptome set. We classified its 51 non-cell line tissues as either survival-crititcal or dispensable. The below table lists the tissues available in GTEx together with their number of biospecimens, whether they were used in our analysis (column Excluded) and whether we considered them critical for survival (column Survival-Critical).

Organ Tissue Available Specimens Excluded Survival-Critical
Adipose Subcutaneous 442 - -
Adipose Visceral (Omentum) 355 - -
Adrenal Gland - 190 - yes
Artery Aorta 299 - -
Artery Coronary 173 - yes
Artery Tibial 441 - -
Bladder - 11 - -
Brain Amygdala 100 - yes
Brain Anterior cingulate cortex (BA24) 121 - yes
Brain Caudate (basal ganglia) 160 - yes
Brain Cerebellar Hemisphere 136 - yes
Brain Cerebellum 173 - yes
Brain Cortex 158 - yes
Brain Frontal Cortex (BA9) 129 - yes
Brain Hippocampus 123 - yes
Brain Hypothalamus 121 - yes
Brain Nucleus accumbens (basal ganglia) 147 - yes
Brain Putamen (basal ganglia) 124 - yes
Brain Spinal cord (cervical c-1) 91 - yes
Brain Substantia nigra 88 - yes
Breast Mammary Tissue 290 - -
Cervix Ectocervix 6 - -
Cervix Endocervix 5 - -
Colon Sigmoid 233 - yes
Colon Transverse 274 - yes
Esophagus Gastroesophageal Junction 244 - yes
Esophagus Mucosa 407 - yes
Esophagus Muscularis 370 - yes
Fallopian Tube - 7 - -
Heart Atrial Appendage 297 - yes
Heart Left Ventricle 303 - yes
Kidney Cortex 45 - yes
Liver - 175 - yes
Lung - 427 - yes
Minor Salivary Gland - 97 - -
Muscle Skeletal 564 - -
Nerve Tibial 414 - -
Ovary - 133 - -
Pancreas - 248 - yes
Pituitary - 183 - -
Prostate - 152 - -
Skin Not Sun Exposed (Suprapubic) 387 - -
Skin Sun Exposed (Lower leg) 473 - -
Small Intestine Terminal Ileum 137 - yes
Spleen - 162 - -
Stomach - 262 - yes
Testis - 259 - -
Thyroid - 446 - -
Uterus - 111 - -
Vagina - 115 - -
Whole Blood - 407 - yes
- EBV-transformed lymphocytes 130 yes -
- Leukemia cell line (CML) 0 yes -
- Transformed fibroblasts 343 yes -