Skip to main content

Expitope 2.0: a tool to assess immunotherapeutic antigens for their potential cross-reactivity against naturally expressed proteins in human tissues



Adoptive immunotherapy offers great potential for treating many types of cancer but its clinical application is hampered by cross-reactive T cell responses in healthy human tissues, representing serious safety risks for patients. We previously developed a computational tool called Expitope for assessing cross-reactivity (CR) of antigens based on tissue-specific gene expression. However, transcript abundance only indirectly indicates protein expression. The recent availability of proteome-wide human protein abundance information now facilitates a more direct approach for CR prediction. Here we present a new version 2.0 of Expitope, which computes all naturally possible epitopes of a peptide sequence and the corresponding CR indices using both protein and transcript abundance levels weighted by a proposed hierarchy of importance of various human tissues.


We tested the tool in two case studies: The first study quantitatively assessed the potential CR of the epitopes used for cancer immunotherapy. The second study evaluated HLA-A*02:01-restricted epitopes obtained from the Immune Epitope Database for different disease groups and demonstrated for the first time that there is a high variation in the background CR depending on the disease state of the host: compared to a healthy individual the CR index is on average two-fold higher for the autoimmune state, and five-fold higher for the cancer state.


The ability to predict potential side effects in normal tissues helps in the development and selection of safer antigens, enabling more successful immunotherapy of cancer and other diseases.

Peer Review reports


The principles of how the immune system can optimally control infections and early stages of cancer underpin the development of immunotherapies. Among these approaches, adoptive transfer of antigen-specific T cells is emerging as a particularly attractive form of immunotherapy to treat patients with more advanced stages of cancer and unresolved infectious diseases. This approach utilizes transfer of tailored antigen-specific immune T cells and provides the possibility of clinically efficient treatment of infectious diseases and human malignancies [1].

One major stumbling block precluding wider application of adoptive immunotherapy is the occurrence of adverse effects of off-target cross-reactivity (CR), which may result in significant, even lethal, toxicity. The cause of toxicity is a hyper-activated T cell response with reactivity directed against normal tissue [2]. Immune CR arises when T cells recognizing a selected target epitope are transferred back to the patient and exhibit recognition of self-epitopes in non-cancerous tissues. On the molecular level this effect is usually the consequence of a high degree of sequence similarity between the target and the self-epitopes, resulting in the binding of a stable self-peptide-MHC complex to the T cell receptor (TCR) and, consequently, cross-activation of unwanted autoimmune T cell responses [3]. Depending on the sequence similarity there can be on-target/off-tumor or off-target recognition. The former is directed against the identical epitope that is also present in a non-cancerous tissue, while the latter is directed against a similar epitope also present in a healthy tissue. The ability to predict the scope and extent of on- and off-target effects can help in selection of safer antigens, and consequently enable more successful immunotherapy treatment [4].

A computational strategy for the prediction of potential peptide-HLA cancer targets and evaluation of the likelihood of off-target toxicity for the targets was developed by Dhanik et al. [5]. The strategy utilizes a sequence-based algorithm similar to the one used in our previous studies [6] and in our current work, but it is not available as a web-service.

We have developed the Expitope server as a tool to assess epitope expression in various tissues (freely accessible at Expitope incorporates the most recent genome-wide information, including protein sequences and protein abundance data across various tissues and cell lines. It enables researchers to screen their epitopes in silico for potential CR in human tissues, before moving their therapeutic candidates into clinical trials.


CR to an immunotherapeutic epitope may arise if a protein normally expressed in healthy cells is cleaved by one of the proteasomes to produce a peptide with an amino acid sequence that is similar to the given epitope. Another prerequisite for CR is the presentation of the natural epitope by major histocompatibility complex class I molecules (MHC-I) in various tissues. We model this process by the method described by Keşmir et al. [7]. To quantitatively assess the natural occurrence of epitopes, we use experimental data on gene expression and abundance of proteins in which the epitopes are present. The methods are described in detail in our previous publication [8] on the iCrossR tool, which has been merged into the current version 2.0 of Expitope. The iCrossR project’s aim was to perform a quantitative characterization study of all MHC-I epitopes listed in the cancer immunotherapy database. A new feature of Expitope 2.0 is the calculation of the tissue-weighted cross-reactivity (CR) indices. Below we test the approach and provide information on the new data sources and a new tissue-weighted CR-index formula.

Material and Implementation

Gene and protein expression data

The previous version 1.0 of Expitope [6] assessed the expression of human antigens based on one combined gene expression database [9] and the Illumina Body Map database [10]. Interestingly, HLA-typing of samples from the Illumina Body Map and Wang et al. [9] showed that the tissues used for expression analysis are most likely derived from the same individual except for seven brain samples [11]. In order to avoid data redundancy with the new Illumina Body Map database, we now only use the brain expression data from Wang et al. [9]. The new version 2.0 of Expitope incorporates three gene expression and four protein abundance datasets (Table 1). It should be noted that in contrast to the PaxDB and Human Proteome Map datasets, which contain ppm values, the Human Protein Atlas data has been generated by immunohistochemistry, which makes the accuracy of the data dependent on the specificity of the antibodies used. The values range from 0 to 3, indicating no detectable expression (0) up to high expression (3).

Table 1 Sources of gene expression and protein abundance data

IEDB datasets

We selected four groups of peptides (Table 2) from the Immune Epitope Database (IEDB) [12], containing a total of 1720 epitopes of 7-25 amino acids in length (Additional file 1: Table S1, Additional file 2: Table S2, Additional file 3: Table S3, Additional file 4: Table S4). The selection for all groups was restricted to the following tags: ’human HLA-A*02:01’, ’Linear Epitopes’, ’Positive Assays only’, ’T cells Assays’, ’MHC ligand Assays’, ’No B-cell assays’, ’Host: Homo Sapiens (Human)’, from which the selection was further restricted for each of the four groups using the tag corresponding to a disease state of the host (column 3 of Table 2).

Table 2 Four epitope groups from the IEDB database

Identification of natural epitopes

Amino acid sequences of epitopes were matched against the RefSeq database [13] of all naturally occurring human protein sequences, including annotated isoforms, downloaded from the National Center for Biotechnology Information (NCBI). The matching procedure yields a list of protein segments, which we call “natural epitopes” (NEs). Potential immunogenicity of each NE was calculated using the formula developed by Keşmir et al. [7], which combines the predicted scores for proteasomal cleavage, TAP affinity and MHC-binding predictions. The quantitative score Q of epitope presentation on MHC-I is defined as:

$$ Q = P_{CL}/\left(A_{TAP}*A_{MHC}\right) $$

where P CL is the proteasomal cleavage probability, while A TAP and A MCH are the IC50-affinities to the transporter molecule associated with antigen processing (TAP) and to the MHC complex, respectively. Lower values for A TAP and A MHC correspond to higher predicted affinities, as IC50-affinity is defined as a dose of peptide that displaces 50% of a competitive ligand.

Calculation of the tissue weighted CR-index

In this version, we modified the CR-index calculation formula [8] to include tissue weighting, reflecting the perceived importance of different tissue types in the human body. For each database, the tissue profile S(t) for a given epitope was calculated as follows:

$$ S(t) = \sum_{k=0}^{K} \left\{ v(k) \cdot \log_{10} \left[ \sum_{i=1}^{M(k)} a(i,t) \right] \right\} $$

where k is the allowed number of mismatches and K is the maximal k; t is the tissue index in a given database of T tissues; i is the running index in the list of matching NEs for each k, and M(k) is the size of the list; v(k) is the normalized mismatch weight, and a(i,t) is the protein or transcript abundance in the tissue t corresponding to the i-th NE. The sum over i includes only the unique NEs that have the scores Q(i) (Equation 1) above a chosen threshold. The normalized mismatch weight is calculated as v(k) = (1/P(k))/ \(\sum \) k(1/P(k)), where P(k) is the probability of finding a random peptide of length l with k mismatches in our protein sequence database of the total length of N=6.5e7 amino acids, P(k) = 1-(1-0.05 l-k ) N-l+1. For example, for a peptide of length 9, the mismatch weights are: v(k=0,1,2,3) = 0.95, 0.0475, 0.0023, 0.0002.

The weighted CR-index is defined as a tissue-weighted average of the tissue profiles S(t):

$$ I_{CR} = \frac{1}{\sum_{t}^{T} w(t)} \sum_{t}^{T} w(t) S(t) $$

where w(t) represents the weight assigned to the tissue type t (Table 3). The I CR index error is obtained as one standard deviation from the mean upon bootstrapping, which involves repeating index calculation 10 times using 90% of randomly subsampled data. The weight values range between 0 and 1, with the weight of 1 corresponding to the most vital organs and systems according to the Sequential Organ Failure Assessment (SOFA) score used to evaluate the condition of patients in Intensive Care Units (ICUs) [14]. The second highest weight of 0.8 is assigned to tissues that belong to vital organs where a failure does not immediately threaten a patient’s life. A weight of 0.5 is assigned to tissues where CR is not necessarily life threatening, but can nevertheless cause severe complications. The second lowest weight of 0.3 refers to tissues and organs that can be surgically removed without major complications. Finally, the weight of 0 was assigned to irrelevant tissues such as testis, where expression of an antigen does not cause an immune response, as well as to the tissues that are only present during pregnancy and other samples that do not correspond to healthy human tissue, e.g. cancer cell lines.

Table 3 Weight values and categorization of tissue types

Consequently, large I CR values may indicate potentially life-threatening CR of the epitope. The higher the number of hits to different NEs that are close in sequence to a therapy peptide, and their total abundance/expression levels in the tissues with high weights, the higher is the probability of CR. Higher thresholds for Q correspond to choosing a higher probability of the selected natural epitope to be immunogenic, while the parameter K controls the sequence similarity: exact match (K=0) for prediction of on-target/off-tumor recognition, and K > 0 for off-target recognition. The values of these parameters can be set by Expitope users. In this work, we chose K=1, i.e. up to one mismatch in amino-acid sequence, and two thresholds for Q: 0.02 corresponding to top 10% immunogenic NEs found for all epitopes in this study, and 1e-4 corresponding to top 50% of the NEs, i.e. top-scored for proteasomal cleavage, TAP transport and MHC-I binding. However, calculation of the indices with the numbers of mismatches K=[0,3] and the combined scores Q=0.02, 1e-4, 1e-5 gave very similar results (Additional file 5: Tables S5-S7; Figure S2).

While a high I CR means that severe complications are expected for a target epitope, its low value hints towards minor or non-life-threatening side effects. An index greater than zero always means that there is some expression present that should be investigated in detail. The index is only an estimate, which does not take into account many patient-specific factors, and therefore should not be used as the sole measure for making decisions. As the tissue classification is not exhaustive and not all organs are completely represented by the tissue types of which they consist, a high expression value in a low rated tissue could correspond to a tissue type not covered, but also present in other more vital organs. Nonetheless, the weighted index offers a short summary of the rather extensive result tables that are produced by Expitope 2.0, and contain individual expression values for each tissue and all NEs. Therefore, the weighted index allows for quick rejection of target epitopes that are likely to cause severe side effects caused by CR.

The I CR indices were calculated with the default parameters (except Q and K) for each peptide and each database using Eq. 3, and were averaged over the seven databases to give the average I CR indices for each peptide. For the plots the I CR indices are averaged for all peptides in each group.

Web server

Expitope 2.0 is a web application that can be easily used by the researchers inexperienced in bioinformatics, especially from the immunotherapy domain. There is no login requirement to the website and user IP addresses are not stored. Multiple clients can connect to the server, and concurrent clients are served one query at a time. The jobs are submitted to high-performance computational infrastructure. The results are displayed once they are ready; alternatively the user can return to the results later, using the session URL. It is also possible to download the results as a spreadsheet to be used with Microsoft Excel or similar software. This allows to sort and filter the results according to individual criteria, e.g. for sorting epitopes by binding affinity predicted by netMHC.

The workflow of Expitope is shown in Fig. 1. The user inputs a peptide sequence and specifies parameters for sequence matching and for the computation of MHC class I binding affinity via the html forms displayed in a web-browser (white). The server performs the search for natural epitopes (NEs) and calculates their Q scores. Computations are performed by the client process at the backend of the server (large gray rectangle). Results are returned to the user in the form of text files and graphical visualizations (dark gray). The user selects a particular database and a plot type for visualization (white). The parameters that can be changed by users in the forms have the following default values: the TAP weight is 0.2, the cleavage threshold is 0.7, the Q score threshold is 1e-4 and the number of mismatches is 2.

Fig. 1
figure 1

Workflow of the Expitope 2.0 web server

Results and discussion

Known cross-reactive epitopes

For the first version of the Expitope web server, the MAGEA3 epitope EVDPIGHLY was tested that had been associated with cross-reactivity caused by the TCR recognizing an epitope with four mismatches derived from titin, which is expressed in heart muscle tissue [6, 15]. We were able to reproduce these findings by using Expitope 2.0 with default the parameters except for allowing up to four mismatches and additionally, the newly added protein databases showed an even clearer result with values of 2.98e+03 ppm (PaxDB) and 2.86e+03 ppm (Human Proteome Map) and the maximum value of 3 for the Human Protein Atlas. Another case of observed cross-reactivity has been a TCR recognizing the MAGEA3/MAGEA9 epitope KVAELVHFL [16]. Expitope 2.0 with the default parameters finds this and all other epitopes from various members of the MAGE family the TCR was able to detect. This includes one epitope of MAGEA12, which was found to be expressed in brain where it led to cross-reactivity. We found expression values of 0.2 FPKM and less but no protein expression for MAGEA12, which is also not contained in the Human Protein Atlas and Human Proteome Map. This demonstrates the importance of taking even small amounts of expression into account when assessing potential cross-reactivity and also comparing the results obtained from all databases, especially for crucial tissues like heart, brain and lung.

Case studies

Cancer immunity peptides

Here we provide an overview of our previous study [8], where we analyzed short (8-15 amino acids) peptide sequences from the Cancer Immunity Peptide Database [17] as well as peptides of viral origin. The CR-index calculation was based only on the PaxDB protein abundance database and without tissue weighting.

The peptide dataset consisted of four groups of currently known human MHC class I epitopes including: mutation antigens displayed by tumor cells (40 peptides, group A), cancer-testis (CT) antigens (67 peptides, group B), differentiation antigens (57 peptides, group C), and overexpressed proteins (94 peptides, group D). In addition, 89 epitopes originating from viral sources (group E) were investigated. When matched exactly, the group of “mutation” antigens produced no hits to the proteins normally expressed in human tissues, since the epitopes of the group have sequences that originated from mutations of normal human protein sequences. The second validation is from the CT antigens, which at small numbers of mismatches (0-1), showed few matches to proteins expressed in the majority of human tissues, with the expected exception of ovary/testis, where multiple hits were found. The hit patterns were very similar for all epitopes of this group. This is exactly as expected, since CT antigens are expressed mostly in these two tissues. In contrast to the results for groups A and B, the antigens of the groups C and D showed more hits, both for exact matches and for high numbers of mismatches. This is also as expected as the proteins containing the epitopes are expressed in a wide variety of normal tissues. Finally, the epitopes originating from the viral sources showed noticeably fewer matches to the human proteins compared to the cancer peptides.

IEDB epitopes

We sought to assess quantitatively the extent of potential “background” CR of the epitopes derived from the host individuals having different disease states - ranging from healthy to cancer. Such background CR is not caused by one single therapy but accumulates due to many factors, including an unknown history of diseases.

The I CR indices of individual epitopes calculated across the seven databases used in this study are highly correlated, since for each database they are obtained by summation of the abundance (or expression) values for the same proteins. There are high correlations between the I CR values computed for the peptides using the three abundance databases as well as between the I CR values derived from the four expression databases (data not shown). Similarly, the correlations between the abundance and expression indices are high (Additional file 5: Figure S1), with the Pearson’s coefficients in the range 0.94-0.96. Averaging of the indices allows one to obtain a more accurate prediction of CR due to increased signal-to-noise ratio, as the databases are derived from different data sources.

Figure 2 shows the I CR indices for the four epitope groups described in Table 2 (group I CR indices before averaging by databases can be found in Additional file 5: Tables S5-S7). The indices for the epitopes computed from 10% top-scoring NEs (Q=0.02, Fig. 2 left) are on average 3-times lower, compared to those from 50% top-scoring NEs (Q=1e-4, Fig. 2 right), corresponding to lower numbers of matching NEs. Higher thresholds for Q correspond to a higher probability of the selected NEs to be immunogenic. It has been reported that the top-scoring 7-10% epitopes identified by the immunogenicity prediction methods have 85% probability of being immunogenic [18]. In this work we have chosen two thresholds of 10% and 50% of sequence matches. The rationale for this choice was to ensure a low amount of false positives in the immunogenicity prediction for the 10% I CR index, and to compare it with the 50% value containing medium to high immunogenic peptides. Two groups - ‘Infectious diseases’ and ‘Healthy’ - have average indices close to zero on both plots, indicating low amounts of cross-reactive epitopes in the critical tissues. The groups ‘Autoimmune diseases’ and ‘Cancer’ exhibit approximately 2- to 5-fold higher average index values compared to the ‘Healthy’ group, in each plot respectively, corresponding to considerably higher presentation level of the cross-reactive peptides in these states.

Fig. 2
figure 2

The I CR indices for the four IEDB peptide groups (Table 2), obtained by averaging over the seven databases listed in Table 1. Q=2e-2 (left), Q=1e-4 (right), with up to one mismatch (K=1). Thick black line: median; gray: the lower and the upper quartiles (25th and 75th percentiles); upper and lower whiskers: highest and lowest values

The interpretation of these results is as follows. The epitopes in the ‘Infectious diseases’ group are derived from non-human organisms rather than from human hosts. Thus, compared to the epitopes from the other three groups, which are of human origin, a lower I CR index is expected, implying low sequence identity to the host and thus a low probability of CR. The slightly elevated index for the ‘Healthy’ group is most likely due to the presence of common pathogens (such as Herpes simplex virus or Epstein-Barr virus) mimicking human sequences, an immune escape strategy known as immune camouflage [19]. A higher I CR for the ‘Autoimmune’ group compared to the ‘Healthy’ group is not surprising, as autoimmunity is a response of the human body’s immune system directed against human proteins overexpressed or aberrantly presented in healthy tissues. For example, multiple sclerosis, the most frequently occurring disease in this group, is due to autoimmunity to the myelin basic protein (MBP), expressed in the tissues of the central nervous system [19]. Other epitopes in this group with very high index values are derived, e.g. from the proteins actin, myosin-9, septin-2 and vimentin, which are normally expressed in various tissues. Normally, peripheral T cells are trained to recognize pathogen-derived epitopes and ignore self-antigens, however some T cells escape this selection and are able to recognize self-antigens, thus initiating an autoimmune response and becoming self-reactive. Consequently with respect to autoimmunity, the term CR is defined as the recognition by T cell TCRs of many different peptide antigens, presented by the HLA of an individual [20], which can also be referred to as cross-recognition.

The significantly higher CR index for the cancer group compared to the other three groups indicates a presence of a high background level of CR when targeting cancers. Cancer epitopes originate either from wild-type proteins overexpressed in tumors, or as a result of cancer-specific mutations in the genes, named neoepitopes. On average, neoepitopes have lower similarity to self-antigens compared to the wild-type cancer epitopes, thus potentially are less cross-reactive. Since T cells with TCRs binding to self-antigens are negatively selected in the thymus, there will generally be a lack of the T cells that can fight tumors, producing overexpressed wild-type proteins. In contrast, the cancers producing neoepitopes can be effectively controlled by the immune system provided that suitable T cells are available. Thus, different types of cancer produce epitopes of varying cross-reactivity, which explains the larger variance seen in Fig. 2 for the cancer group compared to the other groups.

High I CR for the ‘Autoimmune disease’ and ‘Cancer’ groups may also be due to an activated state of the immune system, when immunoproteasomes create larger amounts of immunogenic (in comparison to standard proteasomes) epitopes, including those from the residuals of normal cells killed by the immune system [21]. In addition, disruption of the normal functioning of the ubiquitin proteasome system may result in creation of abnormally presented immunogenic epitopes, leading to many types of disorders, including malignancies, neurodegenerative diseases and systemic autoimmunity [22, 23].

Thus, multiple reasons for a high variability in presented CR epitopes appear to exist depending on the host disease state. This CR, which we tentatively call “background” CR, is independent of any immune therapy. Clearly, a collection of epitopes present in a particular individual is different from our datasets obtained from the IEDB database. Likely, it will include only a subset of the peptides, but a statistical distribution in many patients may exhibit a pattern similar to the one reported in this work. Eventually, it remains to be seen if there can be any interference between the background CR and the CR invoked by a therapy, but both types are important to assess the safety of the therapy.


It is a long-standing dream of many medical practitioners to use the immune system for effective treatment and permanent cure of human disease conditions. With the number of tested and approved immunotherapies growing, evidence of the side effects associated with the current therapies also increased. Consequently, therapy developers require reliable tools for predicting unwanted cross-reactions.

The Expitope web tool for predicting CR of T cell epitopes is based on experimental protein abundance and expression data obtained from a growing number of publicly available databases. We demonstrate its performance for a large number of epitopes detected in the human organism for various cancer types and at various diseases states, ranging from healthy to cancer. The results of our study of Cancer Immunity Peptides [8] showed that the currently known cancer epitopes display a very large CR variability across a range of tissues. Our predictions are in close agreement with the results of several clinical studies, with the CR indices being high in the tissues where actual side effects have been reported, and close to zero for no side-effects. Thus, Expitope enables researchers to assess potential side effects of their selected antigens for therapy and to identify specific human tissues where such side effects could be expected. Since any immunotherapy can cause side effects, we suggest using this tool at both early and late stages of a therapy development process. CR index values calculated by Expitope can serve as an estimate of the amount of potential CR for in silico assessment of immunotherapeutic strategies.

For the first time we demonstrate that there is a high variation in the CR of peptides presented at different disease states of the host: it is on average 2-fold higher for individuals with an autoimmune state and 5-fold higher for individuals with cancer in comparison to individuals in an apparent healthy state. Presumably, a similar background CR may exist prior to an immune therapy, which may differ by the host disease state. Since the human organism negatively pre-selects T cells binding to self-antigens, there will be a small number or no T cells fighting disease tissue cells marked by highly cross-reactive epitopes. Consequently, the similarity of presented epitopes to self-antigens is an obstacle for disease elimination both for the organism itself and for immunotherapy. Thus, therapy developers should consider the possibility of background CR interfering with a therapy.

Availability and requirements

Project name: Expitope 2.0

Project home page:

Operating system(s): Platform independent

Programming language: Java, JavaScript

Other requirements: Web browser

License: None (free to use for academic purposes)

Any restrictions to use by non-academics: None







Human leucocyte antigen


Immune epitope database


Major histocompatibility complex


Natural epitope


Transporter associated with antigen processing


T cell receptor


  1. Ludewig B, (ed).Adoptive Immunotherapy: Methods and Protocols. Methods in molecular medicine, Vol. 109. Totowa: Humana Press; 2005.

    Google Scholar 

  2. Weber JS, Yang JC, Atkins MB, Disis ML. Toxicities of Immunotherapy for the Practitioner. J Clin Oncol. 2015; 33(18):2092–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. Kohm AP, Fuller KG, Miller SD. Mimicking the way to autoimmunity: an evolving theory of sequence and structural homology. Trends Microbiol. 2003; 11(3):101–5.

    Article  CAS  PubMed  Google Scholar 

  4. Maus MV, Fraietta JA, Levine BL, Kalos M, Zhao Y, June CH. Adoptive Immunotherapy for Cancer or Viruses. Annu Rev Immunol. 2014; 32(1):189–225.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Dhanik A, Kirshner JR, MacDonald D, Thurston G, Lin HC, Murphy AJ, Zhang W. In-silico discovery of cancer-specific peptide-HLA complexes for targeted therapy. BMC Bioinformatics. 2016; 17:286.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Haase K, Raffegerst S, Schendel DJ, Frishman D. Expitope: a web server for epitope expression. Bioinformatics. 2015; 31(11):1854–6.

    Article  CAS  PubMed  Google Scholar 

  7. Keşmir C, Nussbaum AK, Schild H, Detours V, Brunak S. Prediction of proteasome cleavage motifs by neural networks. Protein Eng. 2002; 15(4):287–96.

    Article  PubMed  Google Scholar 

  8. Jaravine V, Raffegerst S, Schendel DJ, Frishman D. Assessment of cancer and virus antigens for cross-reactivity in human tissues. Bioinformatics. 2017; 33(1):104–11.

    Article  PubMed  Google Scholar 

  9. Wang ET, Sandberg R, Luo S, Khrebtukova I, Zhang L, Mayr C, Kingsmore SF, Schroth GP, Burge CB. Alternative isoform regulation in human tissue transcriptomes. Nature. 2008; 456(7221):470–6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Barbosa-Morais NL, Irimia M, Pan Q, Xiong HY, Gueroussov S, Lee LJ, Slobodeniuc V, Kutter C, Watt S, Colak R, Kim T, Misquitta-Ali CM, Wilson MD, Kim PM, Odom DT, Frey BJ, Blencowe BJ. The evolutionary landscape of alternative splicing in vertebrate species. Science. 2012; 338(6114):1587–93.

    Article  CAS  PubMed  Google Scholar 

  11. Boegel S, Löwer M, Schäfer M, Bukur T, de Graaf J, Boisguérin V, Türeci O, Diken M, Castle JC, Sahin U. HLA typing from RNA-Seq sequence reads. Genome Med. 2012; 4(12):102.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Vita R, Overton JA, Greenbaum JA, Ponomarenko J, Clark JD, Cantrell JR, Wheeler DK, Gabbard JL, Hix D, Sette A, Peters B. The immune epitope database (IEDB) 3.0. Nucleic Acids Res. 2015; 43(D1):405–12.

    Article  Google Scholar 

  13. Pruitt KD, Tatusova T, Brown GR, Maglott DR. NCBI Reference Sequences (RefSeq): current status, new features and genome annotation policy. Nucleic Acids Res. 2012; 40(D1):130–5.

    Article  Google Scholar 

  14. Vincent JL, Moreno R, Takala J, Willatts S, De Mendonça A, Bruining H, Reinhart CK, Suter PM, Thijs LG. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med. 1996; 22(7):707–10.

    Article  CAS  PubMed  Google Scholar 

  15. Linette GP, Stadtmauer EA, Maus MV, Rapoport AP, Levine BL, Emery L, Litzky L, Bagg A, Carreno BM, Cimino PJ, Binder-Scholl GK, Smethurst DP, Gerry AB, Pumphrey NJ, Bennett AD, Brewer JE, Dukes J, Harper J, Tayton-Martin HK, Jakobsen BK, Hassan NJ, Kalos M, June CH. Cardiovascular toxicity and titin cross-reactivity of affinity-enhanced t cells in myeloma and melanoma. Blood. 2013; 122(6):863–71.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Morgan RA, Chinnasamy N, Abate-Daga D, Gros A, Robbins PF, Zheng Z, Dudley ME, Feldman SA, Yang JC, Sherry RM, Phan GQ, Hughes MS, Kammula US, Miller AD, Hessman CJ, Stewart AA, Restifo NP, Quezado MM, Alimchandani M, Rosenberg AZ, Nath A, Wang T, Bielekova B, Wuest SC, Akula N, McMahon FJ, Wilde S, Mosetter B, Schendel DJ, Laurencot CM, Rosenberg SA. Cancer regression and neurological toxicity following anti-MAGE-A3 TCR gene therapy. J Immunother. 2013; 36(2):133–51.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Vigneron N, Stroobant V, Van den Eynde BJ, van der Bruggen P. Database of T cell-defined human tumor antigens: the 2013 update. Cancer Immun. 2013; 13:15.

    PubMed  PubMed Central  Google Scholar 

  18. Larsen M, Lundegaard C, Lamberth K, Buus S, Brunak S, Lund O, Nielsen M. An integrative approach to CTL epitope prediction: A combined algorithm integrating MHC class I binding, TAP transport efficiency, and proteasomal cleavage predictions. Eur J Immunol. 2005; 35(8):2295–303.

    Article  CAS  PubMed  Google Scholar 

  19. Moise L, Beseme S, Tassone R, Liu R, Kibria F, Terry F, Martin W, De Groot AS. T cell epitope redundancy: cross-conservation of the TCR face between pathogens and self and its implications for vaccines and autoimmunity. Expert Rev Vaccines. 2016; 15(5):607–17.

    Article  CAS  PubMed  Google Scholar 

  20. Kumar A, Delogu F. Dynamical footprint of cross-reactivity in a human autoimmune T-cell receptor. Sci Rep. 2017; 7:42496.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Vigneron N. Human Tumor Antigens and Cancer Immunotherapy. BioMed Res Int. 2015; 2015:1–17.

    Article  Google Scholar 

  22. Nalepa G, Rolfe M, Harper JW. Drug discovery in the ubiquitin–proteasome system. Nat Rev Drug Discov. 2006; 5(7):596–613.

    Article  CAS  PubMed  Google Scholar 

  23. Wang J, Maldonado MA. The ubiquitin-proteasome system and its role in inflammatory and autoimmune diseases. Cell Mol Immunol. 2006; 3(4):255–61.

    CAS  PubMed  Google Scholar 

  24. Wang M, Herrmann CJ, Simonovic M, Szklarczyk D, von Mering C. Version 4.0 of PaxDb: Protein abundance data, integrated across model organisms, tissues, and cell-lines. Proteomics. 2015; 15(18):3163–8.

    Article  CAS  PubMed  Google Scholar 

  25. Petryszak R, Keays M, Tang YA, Fonseca NA, Barrera E, Burdett T, Füllgrabe A, Fuentes AM-P, Jupp S, Koskinen S, Mannion O, Huerta L, Megy K, Snow C, Williams E, Barzine M, Hastings E, Weisser H, Wright J, Jaiswal P, Huber W, Choudhary J, Parkinson HE, Brazma A. Expression Atlas update—an integrated database of gene and protein expression in humans, animals and plants. Nucleic Acids Res. 2016; 44(D1):746–52.

    Article  Google Scholar 

  26. Uhlen M, Fagerberg L, Hallstrom BM, Lindskog C, Oksvold P, Mardinoglu A, Sivertsson A, Kampf C, Sjostedt E, Asplund A, Olsson I, Edlund K, Lundberg E, Navani S, Szigyarto CA-K, Odeberg J, Djureinovic D, Takanen JO, Hober S, Alm T, Edqvist PH, Berling H, Tegel H, Mulder J, Rockberg J, Nilsson P, Schwenk JM, Hamsten M, von Feilitzen K, Forsberg M, Persson L, Johansson F, Zwahlen M, von Heijne G, Nielsen J, Ponten F. Tissue-based map of the human proteome. Science. 2015; 347(6220):1260419.

    Article  PubMed  Google Scholar 

  27. Kim MS, Pinto SM, Getnet D, Nirujogi RS, Manda SS, Chaerkady R, Madugundu AK, Kelkar DS, Isserlin R, Jain S, Thomas JK, Muthusamy B, Leal-Rojas P, Kumar P, Sahasrabuddhe NA, Balakrishnan L, Advani J, George B, Renuse S, Selvan LDN, Patil AH, Nanjappa V, Radhakrishnan A, Prasad S, Subbannayya T, Raju R, Kumar M, Sreenivasamurthy SK, Marimuthu A, Sathe GJ, Chavan S, Datta KK, Subbannayya Y, Sahu A, Yelamanchi SD, Jayaram S, Rajagopalan P, Sharma J, Murthy KR, Syed N, Goel R, Khan AA, Ahmad S, Dey G, Mudgal K, Chatterjee A, Huang TC, Zhong J, Wu X, Shaw PG, Freed D, Zahari MS, Mukherjee KK, Shankar S, Mahadevan A, Lam H, Mitchell CJ, Shankar SK, Satishchandra P, Schroeder JT, Sirdeshmukh R, Maitra A, Leach SD, Drake CG, Halushka MK, Prasad TSK, Hruban RH, Kerr CL, Bader GD, Iacobuzio-Donahue CA, Gowda H, Pandey A. A draft map of the human proteome. Nature. 2014; 509(7502):575–81.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  28. Lonsdale J, Thomas J, Salvatore M, Phillips R, Lo E, Shad S, Hasz R, Walters G, Garcia F, Young N, Foster B, Moser M, Karasik E, Gillard B, Ramsey K, Sullivan S, Bridge J, Magazine H, Syron J, Fleming J, Siminoff L, Traino H, Mosavel M, Barker L, Jewell S, Rohrer D, Maxim D, Filkins D, Harbach P, Cortadillo E, Berghuis B, Turner L, Hudson E, Feenstra K, Sobin L, Robb J, Branton P, Korzeniewski G, Shive C, Tabor D, Qi L, Groch K, Nampally S, Buia S, Zimmerman A, Smith A, Burges R, Robinson K, Valentino K, Bradbury D, Cosentino M, Diaz-Mayoral N, Kennedy M, Engel T, Williams P, Erickson K, Ardlie K, Winckler W, Getz G, DeLuca D, MacArthur D, Kellis M, Thomson A, Young T, Gelfand E, Donovan M, Meng Y, Grant G, Mash D, Marcus Y, Basile M, Liu J, Zhu J, Tu Z, Cox NJ, Nicolae DL, Gamazon ER, Im HK, Konkashbaev A, Pritchard J, Stevens M, Flutre T, Wen X, Dermitzakis ET, Lappalainen T, Guigo R, Monlong J, Sammeth M, Koller D, Battle A, Mostafavi S, McCarthy M, Rivas M, Maller J, Rusyn I, Nobel A, Wright F, Shabalin A, Feolo M, Sharopova N, Sturcke A, Paschal J, Anderson JM, Wilder EL, Derr LK, Green ED, Struewing JP, Temple G, Volpi S, Boyer JT, Thomson EJ, Guyer MS, Ng C, Abdallah A, Colantuoni D, Insel TR, Koester SE, Little AR, Bender PK, Lehner T, Yao Y, Compton CC, Vaught JB, Sawyer S, Lockhart NC, Demchok J, Moore HF. The Genotype-Tissue Expression (GTEx) project. Nat Genet. 2013; 45(6):580–5.

    Article  CAS  Google Scholar 

  29. Forrest ARR, Kawaji H, Rehli M, Baillie JK, de Hoon MJL, Haberle V, Lassmann T, Kulakovskiy IV, Lizio M, Itoh M, Andersson R, Mungall CJ, Meehan TF, Schmeier S, Bertin N, Jørgensen M, Dimont E, Arner E, Schmidl C, Schaefer U, Medvedeva YA, Plessy C, Vitezic M, Severin J, Semple CA, Ishizu Y, Young RS, Francescatto M, Alam I, Albanese D, Altschuler GM, Arakawa T, Archer JAC, Arner P, Babina M, Rennie S, et al.A promoter-level mammalian expression atlas. Nature. 2014; 507(7493):462–70.

    Article  CAS  PubMed  Google Scholar 

  30. Prete M, Dammacco R, Fatone MC, Racanelli V. Autoimmune uveitis: clinical, pathogenetic, and therapeutic features. Clin Exp Med. 2016; 16(2):125–36.

    Article  CAS  PubMed  Google Scholar 

Download references





Availability of data and materials

The datasets generated and/or analysed during the current study are available at

Author information

Authors and Affiliations



VJ further developed the algorithm and the web implementation. AM added the tissue scoring function. VJ and AM analyzed the data and wrote the manuscript. SR, DJS and DF conducted the project and edited the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Dmitrij Frishman.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

VJ, AM and SR are employees of Medigene Immunotherapies GmbH/Medigene AG. DJS is Managing Director of Medigene Immunotherapies GmbH and CEO/CSO of Medigene AG.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Additional files

Additional file 1

TableS1_InfectiousDisease. Comma-separated table containing Table S1. (CSV 80 kb)

Additional file 2

TableS2_Healthy. Comma-separated table containing Table S2. (CSV 67 kb)

Additional file 3

TableS3_AutoimmuneDisease. Comma-separated table containing Table S3. (CSV 20 kb)

Additional file 4

TableS4_Cancer. Comma-separated table containing Table S4. (CSV 67 kb)

Additional file 5

Suppl-Material. Microsoft Word file containing Figures S1 and S2 and Tables S5-S7. (DOCX 191 kb)

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Jaravine, V., Mösch, A., Raffegerst, S. et al. Expitope 2.0: a tool to assess immunotherapeutic antigens for their potential cross-reactivity against naturally expressed proteins in human tissues. BMC Cancer 17, 892 (2017).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: