- Open Access
AbMiner: A bioinformatic resource on available monoclonal antibodies and corresponding gene identifiers for genomic, proteomic, and immunologic studies
© Major et al; licensee BioMed Central Ltd. 2006
- Received: 21 October 2005
- Accepted: 06 April 2006
- Published: 06 April 2006
Monoclonal antibodies are used extensively throughout the biomedical sciences for detection of antigens, either in vitro or in vivo. We, for example, have used them for quantitation of proteins on "reverse-phase" protein lysate arrays. For those studies, we quality-controlled > 600 available monoclonal antibodies and also needed to develop precise information on the genes that encode their antigens. Translation among the various protein and gene identifier types proved non-trivial because of one-to-many and many-to-one relationships. To organize the antibody, protein, and gene information, we initially developed a relational database in Filemaker for our own use. When it became apparent that the information would be useful to many other researchers faced with the need to choose or characterize antibodies, we developed it further as AbMiner, a fully relational web-based database under MySQL, programmed in Java.
AbMiner is a user-friendly, web-based relational database of information on > 600 commercially available antibodies that we validated by Western blot for protein microarray studies. It includes many types of information on the antibody, the immunogen, the vendor, the antigen, and the antigen's gene. Multiple gene and protein identifier types provide links to corresponding entries in a variety of other public databases, including resources for phosphorylation-specific antibodies. AbMiner also includes our quality-control data against a pool of 60 diverse cancer cell types (the NCI-60) and also protein expression levels for the NCI-60 cells measured using our high-density "reverse-phase" protein lysate microarrays for a selection of the listed antibodies. Some other available database resources give information on antibody specificity for one or a couple of cell types. In contrast, the data in AbMiner indicate specificity with respect to the antigens in a pool of 60 diverse cell types from nine different tissues of origin.
AbMiner is a relational database that provides extensive information from our own laboratory and other sources on more than 600 available antibodies and the genes that encode the antibodies' antigens. The data will be made freely available at http://discover.nci.nih.gov/abminer
- Relational Database
- Gene Identifier
- Antibody Reagent
- Lysate Array
- Spotted cDNA Array
Antibodies are used as tools throughout biomedical science, and they are, increasingly, being incorporated into clinical practice in such specialties as rheumatology, oncology, and infectious diseases . They are also finding more and more application in the new high-throughput biotechnologies such as antibody and protein lysate microarrays [2–8]. As a consequence of that increased prominence and range of application, antibody reagents (particularly monoclonals) are being made available to the researcher commercially in increasing numbers. However, some of them do not have the right affinity, specificity, or other characteristics for a particular application, creating a problem and, often, wasted effort for end-users .
That was the case when our laboratory began the project that motivated us to develop AbMiner: 'reverse-phase' protein lysate microarray profiling of the 60 human cancer cell lines (the NCI-60) used since 1990 by the U.S. National Cancer Institute's Developmental Therapeutics Program to screen > 100,000 chemical compounds (plus natural products) for anticancer activity [9, 10]. In 2001, Paweletz, et al.  introduced 'reverse phase' protein lysate microarrays (henceforth, called 'lysate arrays' here), in combination with laser capture microdissection and robotic spotting technology. For the NCI-60 project, we  then developed higher density lysate arrays that incorporated all 60 cell lines plus controls, each at 10 serial two-fold dilutions to achieve wide dynamic range and good reproducibility (17% coefficient of variation) in profiling of protein levels across the cell types. Antibodies were used to quantify protein on the arrays using a Catalyzed Signal Amplification method (DAKO Cytomation, Carpenteria, CA, USA). We obtained more than 600 commercially available monoclonal antibodies to find ones suitable for the purpose. Before application to the arrays, we screened the antibodies by Western blot against a pool of the NCI-60 lysates (equal amounts from each cell type). Since the pool included cancer cell lines from 9 different tissues of origin, it served as an extensive (though not exhaustive) sampling of human protein antigens.
Data fields in AbMiner. Fields that can be searched using AbMiner's Advanced Search function are indicated.
Fields in AbMiner
Molecular Weight Range/KDa
UniGene Cluster Id
Entrez Gene Id
Recommended start dilution
In addition, AbMiner provides a connection to other 'omic' data  by matching each antibody with the target antigen's corresponding DNA and RNA identifiers. We were initially motivated to translate antibody names to gene symbols because of studies in which we were correlating protein and mRNA expression from microarrays for biomarker discovery . The gene information in AbMiner was later expanded to include a variety of genomic and proteomic identifiers. Using our MatchMiner program package , a tool for batch-translation of identifiers, we matched each antibody to its antigen's gene symbol. The corresponding gene name, UniGene cluster ID, LocusLink ID, and RefSeq were then identified using MatchMiner and other tools [15–22], with careful manual curation. AbMiner provides link-outs from gene identifiers to the corresponding entries in various public resources (LocusLink, GeneCards, etc.), as well as to other antibody databases. By providing a means to search by DNA, RNA, or protein, AbMiner facilitates the integration of genomic, transcriptomic, and proteomic information.
Monoclonal antibodies directed against protein antigens were obtained from many different commercial sources (listed in the AbMiner program itself), with no particular selection criteria except that antibodies directed against very small (~ 10 kDa) or very large (> 350 kDa) proteins were excluded. Species recognized included human, mouse, rat, dog, chicken, frog, and others. Each vial of antibody was assigned a unique AbMiner identification code number so that screening of the particular vial could be tracked. Pertinent information indicated in the schema in Table 1 was also recorded.
The Western blot results were classified as follows: (a) single band: one predominant band, at the expected molecular weight; (b) multiple bands: extra bands remaining with a 5-second or longer film exposure; (c) wrong molecular weight: predominant band or bands at unexpected molecular weight(s); and (d) no band. Antibodies were Western-blotted up to three or four times if required to obtain clear results.
We focused on Western blot analysis and designed the screening process as we did because of the specific requirements for application of antibodies on reverse-phase lysate arrays. Antibodies that recognize an epitope from more than one protein (or isoform) can be used for detection and quantitation of proteins on a Western blot as long as any extraneous bands have different effective molecular weights and would show up as separate bands. The lysate array, in contrast, is effectively a multiplexed dot blot; the signal from each spot on the array is the summation of specific and non-specific binding of the antibody. Therefore, for the lysate arrays we used only screened antibodies that produced a single predominant band by Western at the expected molecular weight. However, antibodies that produced multiple bands may still be useful for other applications, so information on them is retained in AbMiner. Other types of quality control data (for example, based on immunoprecipitation, immunohistochemistry, or flow cytometry) may be most pertinent to other types of applications. AbMiner is extensible in that data fields can be added to accommodate and present such information.
AbMiner's Gene Information database provides translation among different data platforms and makes it possible for the user to search by proteomic, transcriptomic, or genomic identifiers. To find the intersection between data sets from different platforms – such as cDNA [25, 26] and oligonucleotide microarrays – one generally must translate from one type of unique identifier to another . Finding an antibody that corresponds to a particular gene can be problematic because many commercially available antibodies do not have unique, universally used names that represent the target gene product.
System design and implementation
AbMiner is a relational database comprised of two major components: (i) a data entry module constructed using FileMaker Pro5.0™ (Santa Clara, CA USA) and used by our team for data entry as well as for detailed tracking of the antibody validation process.
(ii) a web application for sharing the various types of information on antibodies, antigens, and genes with the research community. The web application, written principally in Java, leverages a variety of available resources: MySQL as the database engine; Hibernate to map the objects into the database; JSP, Struts, and Tiles to render the user interface; and JUnit and HTTPUnit for testing individual programming units and the overall system. In addition to providing the web user interface, we have defined a simple HTTP specification that facilitates linkage of other applications into AbMiner. The web application was constructed under the "Agile Development" paradigm, which encourages close, iterative interaction between user/tester/motivators of the package (biologists) and software engineers . That interaction, and the continuing revision of specs that the agile process encourages, ensure that AbMiner will serve broad needs of biological researchers. It has been received enthusiastically in extensive beta-testing.
Both components of AbMiner have the same underlying model, which includes three main modules: Antibody, Screening, and Gene. AbMiner uses a relational database approach to manage the complex relationships among those elements. The relationships are generally not one-to-one. For example, a given gene often codes for different splice variants, which may or may not be recognized by the same antibody. Conversely, multiple antibodies from different vendors or hybridoma clones may target a protein encoded by a single gene. An additional complication is that, because of the continuing re-annotation of the human genome, some identifiers are not unique or constant. That dynamic process is exemplified by retired or relocated UniGene clusters that can sometimes result in more than one UniGene Cluster ID or LocusID entry for the same gene. By constructing AbMiner as a relational database, we have been able to organize and update those one-to-many, many-to-one, and many-to-many relationships.
Currently, AbMiner is populated with screening data generated by our own laboratory, but we plan to incorporate data from other studies and repositories when available. We also plan to put data entry pages on the web component for input of screening information from other investigators in the research community who wish to contribute (with appropriate attribution)
The identifiers described in the last section perform an additional function by serving as link-outs to their respective entries in LocusLink, UniGene, GeneCards, Entrez's RefSeq, and our MedMiner program. AbMiner also provides links to the Mammalian Phosphorylation Resource (MPR) , a web site that contains sequence information for phosphorylation sites recognized with specificity by commercial antibodies, and links to the Clinical Proteomics Databank , which provides a list of phospho-specific antibodies tested and used in the Clinical Proteomics Program of the NCI. Other public and commercial antibody databases, such as the Antibody Resource page and Abcam , are also linked. Finally, AbMiner will serve as the central public database for an antibody repository planned by the NCI Center for Cancer Research (CCR).
AbMiner applied to molecular biomarker identification
As already noted, development of AbMiner was motivated by our need to organize information on antibodies for lysate array studies, and it has proved itself an almost indispensable tool in that respect. Particularly important is the information on the correspondence between antibody names, antigen names, and the variety of gene identifier types. We were able, for example, to address the question of similarities and differences between mRNA and protein expression profiles across the NCI-60 . Identifiers of proteins quantitated on lysate arrays were matched with identifiers of transcripts assessed on spotted cDNA arrays (i.e., Image Clone Ids) and Affymetrix oligonucleotide arrays (i.e., Affymetrix Ids) using MatchMiner  and AbMiner. A central, unexpected finding was that cell-structure-related proteins showed higher correlation between protein and mRNA levels across the 60 cell lines than did non-cell-structure-related proteins . Using the annotations and translation capabilities in AbMiner, those analyses have now been extended to 89 proteins detected on the lysate arrays by 154 different antibodies (Shankavaram et al. manuscript in preparation).
We have also applied the resources of AbMiner to the identification of molecular biomarkers at the protein level. For one such study, we developed a multi-step "integromic" protocol [14, 30], which included: (a) identification of candidate markers using cDNA microarrays; (b) re-sequencing of candidate clones; (c) corroboration of the candidates' expression patterns from the cDNA microarray using Affymetrix oligonucleotide chips; (d) protein expression analysis using reverse-phase protein lysate arrays; and (e) prospective validation of candidate biomarkers on tissue microarrays consisting of hundreds of tumor samples. With that algorithm we identified villin and moesin as molecular markers that distinguish between colon and ovarian adenocarcinomas. Those cancer types can be difficult to distinguish in a few percent of metastatic or disseminated lesions in the abdomen, and the differential diagnosis is important because it determines what drugs will be used for therapy. Our protocol was successful in that case, but it depended on the availability and effective screening of quality antibodies for identification of diagnostic markers at the protein level on the lysate and tissue arrays. AbMiner gene identifiers will help other investigators in similar searches for molecular markers at the protein level, even when the search has begun with genomic databases. Because AbMiner provides extensive information for over 600 validated antibodies, the transcriptional signature of a gene can often be corroborated directly at the protein level.
Identifiers matched to AbMiner antibodies and the number of antibodies with each identifier.
AbMiner Gene Information Records
Number of antibodies
Total antibodies in collection
Name (gene symbol)
Accession # (Refseq)
Unigene Cluster ID
Antibodies in AbMiner with matching UniGene Cluster IDs in four microarray platforms
Number of antibodies in AbMiner that match microarray identifiers
Number of antobodies
Antibodies with UniGene IDs
Oligo 6.8 K
Comparison with other antibody databases
AbMiner is certainly not the most comprehensive in terms of numbers of antigens or antibodies covered. In that regard, the classic source is Linscott's Directory . The Advanced Type Culture Collection (ATCC)  keeps listings of large numbers of their available cell lines, including hybridomas, and the Antibody Resource Page  and Abcam  provide useful databases on antibodies. AbMiner includes links to all of those sources. A number of companies provide databases of the antibodies they sell, but those will not be reviewed here. AfCS Signaling Gateway provides  provides information on 138 proteins (principally in the signaling pathways) and antibodies against them. Western Blot quality control information, generally on one or a few cell lines, is included. Exactantigen  provides gene-specific and species-specific information on antibodies, with links to manufacturers' data sheets. The useful Human Protein Atlas [40, 41] features immunohistochemical images for a variety of newly generated and other antibodies, complementing the focus of AbMiner. There are also a number of specialized antibody collections (e.g., on 3-D structures of antibodies or on neurological or HIV-related reagents) [42, 43], but none that we have seen present ranges of information similar to that of AbMiner. It would be well beyond the scope of this article to review those databases, but a number of them are described, with outlinks, by Linscott's. AbMiner's database will continue to expand, but not with the intention of competing with Linscott's in coverage.
Overall, to the best of our knowledge, none of the other sources have the range of information types on the antibodies, the antigens, the vendors, and the antigen's genes that AbMiner does, and none of them give the type of multiple-tissue Western blot specificity data or protein microarray data that are compiled in AbMiner. Insofar as we have found, any Western blot results given in the other databases had been obtained against single cell types. The quality control criteria represented in AbMiner are stricter and more comprehensive in that we have validated the antibodies by Western blot against a pool representing a wide range of cancer cell lines from nine different organs of origin and from different cell lineages. Non-specificities showed up in that more rigorous testing when they didn't in testing against individual cell types.
AbMiner has unique relational characteristics for dealing with the one-to-many, many-to-one, and many-to-many relationships among antibody reagents, their antigens, and the genes of those antigens. Through the use of MatchMiner , supplemented by manual curation from additional bioinformatic resources, AbMiner gives a useful range of gene identifier types not otherwise easy for the casual user to find. The Antibody Resource Page provides a listing of "databases/software" on immunological reagents , but none of those listed have major overlap with AbMiner in terms of their program and search capabilities. We are currently using the structure of AbMiner as template for an analogous database on siRNA reagents.
We developed AbMiner as we did to provide the type of information needed for "integromic" [30, 44] studies of the type described above for biomarker identification – that is, for the integration of different types of molecular data at the DNA, RNA, protein, and functional levels. But the program is also being found useful (in beta-testing) by researchers with simpler aims: e.g., those who simply want to find the right antibody for an assay.
AbMiner is freely accessible to both public and private sector users at http://discover.nci.nih.gov/abminer. Also available there for batch downloading are the quality control results and lysate array data for screened antibodies. They will be updated as new antibodies are tested . Also available there is a detailed protocol for the Western blot screening. Gene Information FILES will be updated regularly. As a Java implementation, AbMiner is browser-, operating system-, and platform-independent.
Authors greatly appreciate Steve Shaw, Emanuel Petricoin, and Lance Liotta for providing antibody database links. This work was supported by the Intramural Research Program of the NIH, National Cancer Institute, Center for Cancer Research.
- Stockwin LH, Holmes S: Antibodies as therapeutic agents: vive la renaissance! Expert Opin Biol Ther 2003, 3: 1133–1152. 10.1517/14712518.104.22.1683View ArticlePubMedGoogle Scholar
- Mendoza LG, McQuary P, Mongan A, Gangadharan R, Brignac S, Eggers M: High-throughput microarray-based enzyme-linked immunosorbent assay (ELISA). Biotechniques 1999, 27: 778–80, 782–6, 788.PubMedGoogle Scholar
- MacBeath G, Schreiber SL: Printing proteins as microarrays for high-throughput function determination. Science 2000, 289: 1760–1763.PubMedGoogle Scholar
- Zhu H, Klemic JF, Chang S, Bertone P, Casamayor A, Klemic KG, Smith D, Gerstein M, Reed MA, Snyder M: Analysis of yeast protein kinases using protein chips. Nat Genet 2000, 26: 283–289. 10.1038/81576View ArticlePubMedGoogle Scholar
- Haab BB, Dunham MJ, Brown PO: Protein microarrays for highly parallel detection and quantitation of specific proteins and antibodies in complex solutions. Genome Biol 2001, 2: RESEARCH0004. 10.1186/gb-2001-2-2-research0004PubMed CentralView ArticlePubMedGoogle Scholar
- Wu G, Datar RH, Hansen KM, Thundat T, Cote RJ, Majumdar A: Bioassay of prostate-specific antigen (PSA) using microcantilevers. Nat Biotechnol 2001, 19: 856–860. 10.1038/nbt0901-856View ArticlePubMedGoogle Scholar
- Houseman BT, Huh JH, Kron SJ, Mrksich M: Peptide chips for the quantitative evaluation of protein kinase activity. Nat Biotechnol 2002, 20: 270–274. 10.1038/nbt0302-270View ArticlePubMedGoogle Scholar
- Weiler T, Sauder P, Cheng K, Ens W, Standing K, Wilkins JA: A proteomics-based approach for monoclonal antibody characterization. Anal Biochem 2003, 321: 217–225. 10.1016/S0003-2697(03)00469-XView ArticlePubMedGoogle Scholar
- Paull KD, Shoemaker RH, Hodes L, Monks A, Scudiero DA, Rubinstein L, Plowman J, Boyd MR: Display and analysis of patterns of differential activity of drugs against human tumor cell lines: development of mean graph and COMPARE algorithm. J Natl Cancer Inst 1989, 81: 1088–1092.View ArticlePubMedGoogle Scholar
- Weinstein JN, Myers TG, O'Connor PM, Friend SH, Fornace AJJ, Kohn KW, Fojo T, Bates SE, Rubinstein LV, Anderson NL, Buolamwini JK, van Osdol WW, Monks AP, Scudiero DA, Sausville EA, Zaharevitz DW, Bunow B, Viswanadhan VN, Johnson GS, Wittes RE, Paull KD: An information-intensive approach to the molecular pharmacology of cancer. Science 1997, 275: 343–349. 10.1126/science.275.5298.343View ArticlePubMedGoogle Scholar
- Paweletz CP, Charboneau L, Bichsel VE, Simone NL, Chen T, Gillespie JW, Emmert-Buck MR, Roth MJ, Petricoin IE, Liotta LA: Reverse phase protein microarrays which capture disease progression show activation of pro-survival pathways at the cancer invasion front. Oncogene 2001, 20: 1981–1989. 10.1038/sj.onc.1204265View ArticlePubMedGoogle Scholar
- Nishizuka S, Charboneau L, Young L, Major S, Reinhold WC, Waltham M, Kouros-Mehr H, Bussey KJ, Lee JK, Espina V, Munson PJ, Petricoin E, Liotta LA, Weinstein JN: Proteomic profiling of the NCI-60 cancer cell lines using new high-density reverse-phase lysate microarrays. Proc Natl Acad Sci U S A 2003, 100: 14229–14234. 10.1073/pnas.2331323100PubMed CentralView ArticlePubMedGoogle Scholar
- Weinstein JN: 'Omic' and hypothesis-driven research in the molecular pharmacology of cancer. Curr Opin Pharmacol 2002, 2: 361–365. 10.1016/S1471-4892(02)00185-6View ArticlePubMedGoogle Scholar
- Nishizuka S, Chen ST, Gwadry FG, Alexander J, Major SM, Scherf U, Reinhold WC, Waltham M, Charboneau L, Young L, Bussey KJ, Kim S, Lababidi S, Lee JK, Pittaluga S, Scudiero DA, Sausville EA, Munson PJ, Petricoin EF, Liotta LA, Hewitt SM, Raffeld M, Weinstein JN: Diagnostic markers that distinguish colon and ovarian adenocarcinomas: identification by genomic, proteomic, and tissue array profiling. Cancer Res 2003, 63: 5243–5250.PubMedGoogle Scholar
- Bussey KJ, Kane D, Sunshine M, Narasimhan S, Nishizuka S, Reinhold WC, Zeeberg B, Ajay W, Weinstein JN: MatchMiner: a tool for batch navigation among gene and gene product identifiers. Genome Biol 2003, 4: R27. 10.1186/gb-2003-4-4-r27PubMed CentralView ArticlePubMedGoogle Scholar
- Anderson NL, Esquer-Blasco R, Hofmann JP, Anderson NG: A two-dimensional gel database of rat liver proteins useful in gene regulation and drug effects studies. Electrophoresis 1991, 12: 907–930. 10.1002/elps.1150121110View ArticlePubMedGoogle Scholar
- Myers TG, Anderson NL, Waltham M, Li G, Buolamwini JK, Scudiero DA, Paull KD, Sausville EA, Weinstein JN: A protein expression database for the molecular pharmacology of cancer. Electrophoresis 1997, 18: 647–653. 10.1002/elps.1150180351View ArticlePubMedGoogle Scholar
- Ross DT, Scherf U, Eisen MB, Perou CM, Rees C, Spellman P, Iyer V, Jeffrey SS, Van de Rijn M, Waltham M, Pergamenschikov A, Lee JC, Lashkari D, Shalon D, Myers TG, Weinstein JN, Botstein D, Brown PO: Systematic variation in gene expression patterns in human cancer cell lines. Nat Genet 2000, 24: 227–235. 10.1038/73432View ArticlePubMedGoogle Scholar
- Scherf U, Ross DT, Waltham M, Smith LH, Lee JK, Tanabe L, Kohn KW, Reinhold WC, Myers TG, Andrews DT, Scudiero DA, Eisen MB, Sausville EA, Pommier Y, Botstein D, Brown PO, Weinstein JN: A gene expression database for the molecular pharmacology of cancer. Nat Genet 2000, 24: 236–244. 10.1038/73439View ArticlePubMedGoogle Scholar
- Staunton JE, Slonim DK, Coller HA, Tamayo P, Angelo MJ, Park J, Scherf U, Lee JK, Reinhold WO, Weinstein JN, Mesirov JP, Lander ES, Golub TR: Chemosensitivity prediction by transcriptional profiling. Proc Natl Acad Sci U S A 2001, 98: 10787–10792. 10.1073/pnas.191368598PubMed CentralView ArticlePubMedGoogle Scholar
- Group GB[http://discover.nci.nih.gov]
- Kane D: Introducing Agile Development into Bioinformatics: An Experience Report: ; Salt Lake City, USA. ; 2003.Google Scholar
- Databank CP[http://home.ccr.cancer.gov/ncifdaproteomics/pmicroarray.asp]
- Page AR[http://www.antibodyresource.com/]
- Gateway ACSS[http://www.signaling-gateway.org/data/antibody/cgi-bin/targets.cgi]
- Atlas HP[http://www.hpr.se/index.php]
- Uhlen M, Bjorling E, Agaton C, Szigyarto CA, Amini B, Andersen E, Andersson AC, Angelidou P, Asplund A, Asplund C, Berglund L, Bergstrom K, Brumer H, Cerjan D, Ekstrom M, Elobeid A, Eriksson C, Fagerberg L, Falk R, Fall J, Forsberg M, Bjorklund MG, Gumbel K, Halimi A, Hallin I, Hamsten C, Hansson M, Hedhammar M, Hercules G, Kampf C, Larsson K, Lindskog M, Lodewyckx W, Lund J, Lundeberg J, Magnusson K, Malm E, Nilsson P, Odling J, Oksvold P, Olsson I, Oster E, Ottosson J, Paavilainen L, Persson A, Rimini R, Rockberg J, Runeson M, Sivertsson A, Skollermo A, Steen J, Stenvall M, Sterky F, Stromberg S, Sundberg M, Tegel H, Tourle S, Wahlund E, Walden A, Wan J, Wernerus H, Westberg J, Wester K, Wrethagen U, Xu LL, Hober S, Ponten F: A Human Protein Atlas for Normal and Cancer Tissues Based on Antibody Proteomics. Mol Cell Proteomics 2005, 4: 1920–1932. 10.1074/mcp.M500279-MCP200View ArticlePubMedGoogle Scholar
- Instituto de biotechnologia UNAM[http://www.ibt.unam.mx/vir/structure/structures.html]
- HIV Molecular immunology database[http://www.hiv.lanl.gov/content/immunology/ab_search]
- Weinstein JN, Pommier Y: Transcriptomic analysis of the NCI-60 cancer cell lines. Comptes Rendus Biology 2003, 326: 909–920. 10.1016/j.crvi.2003.08.005View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.