doi: 10.1111/cea.12569 Clinical & Experimental Allergy, 45, 1259–1261 RESEARCH LETTER © 2015 John Wiley & Sons Ltd DAAB: a manually curated database of allergy and asthma biomarkers G. Sircar 1, *, B. Saha 1, *, T. Jana 2, *, A. Dasgupta 3 , S. Gupta Bhattacharya 1 and S. Saha 2 1 Division of Plant Biology, Bose Institute, Kolkata, India, 2 Bioinformatics Center, Bose Institute, Kolkata, India and 3 Department of Medicine, BR Singh Hospital and Centre for Medical Education and Research, Kolkata, India Allergy and asthma have reached a pandemic dimen- sion, and search for good biomarkers for this disease is of considerable clinical interests [1]. The absence of accurate biomarker for allergy has rendered diagnosis and phenotyping of this disease largely inexplicable [2]. Currently, the mainstay of allergy treatment is symp- tomatic rather than targeted, using anti-inflammatories and steroidal drugs, which themselves cause adverse side effects [3]. Furthermore, certain non-allergic disor- ders such as neutrophilic asthma, chronic obstructive pulmonary disease (COPD) and skin irritation are often misdiagnosed as allergic diseases [4, 5]. Therefore, allergy phenotype-specific biomarkers for accurate diagnosis, early prediction and targeted drug therapy are of prime importance [6]. Biomarker database is available for infectious diseases [7], but not for allergy and asthma. Therefore, we felt there is an urgent need to compile published research work for this disease, so that the potential biomarkers of this disease can be available on a single platform. Database of Allergy and Asthma Biomarkers (DAAB) is a manually curated repository of biomarkers of different types of allergic diseases. We referred biomarkers as active genes/pro- teins, which are found to be statistically significant in differential expression profiling and considerably mod- ulated in allergy and asthma diseases. About 2154 entries have been compiled by text mining of PubMed abstracts followed by detailed manual curation, of which 1022 entries from genomics, 419 from proteo- mics, 16 entries from epigenetics and 210 entries from other low-throughput studies. DAAB contains informa- tion on identified biomarker accession numbers (NCBI and UniProt), along with experimental approaches (techniques, OMICS), disease phenotype and tissue sam- ples types. In addition, it provides link to PubMed for reference, Gene Expression Omnibus (GEO) and Proteo- mics IDEntifications (PRIDE) databases for archiving the microarray and mass spectrometry data sets, respec- tively, Drug Bank for drug target and monoclonal anti- body, if available, for potential therapy or further downstream validation. The users can query through user-friendly search page and browse the data using alphabetical order of the biomarker gene symbols. The data can also be downloaded in flat format. DAAB is freely accessible and contains 1200 unique biomarkers. The entire data in DAAB have been organized, and users can retrieve and analyse these data in three ways as shown in Fig. 1. First, it allows browse option to the users. The database can be browsed using alphabetical list of gene symbols. For example, on selecting ‘A’ (See Fig. 2a), a list of 168 biomarkers having their gene symbols starting with ‘A’ will appear as output (See Fig. 2b). In addition, the data can be browsed using four different experimental approaches such as Genom- ics, Proteomics, Epigenetics and Others. For example, hitting on ‘Genomics’ a list of 1252 biomarkers, which have been obtained from various genomic platforms, will appear as an output. Second, it allows search option using keyword to the users. The keyword search can be restricted by filtering data through either a gene symbol or tissue sample utilized in an experiment or OMICS approach used, by selecting from a drop-down menu. Alternatively, any keyword can also be searched by specifying ‘All’ option from this menu. This ‘search’ helps a user to explore any experiment that has been carried out previously related to a particular keyword as well as its relevance to any allergic diseases. For example, searching with gene symbol ‘Muc5ac’ as a keyword within ‘All’ data (See Fig. 2c) will show 10 records (See Fig. 2d), which implies that ‘Muc5ac’ gene was found to be reported ten times in ten different experiments cited in DAAB. Third, it allows BLAST option torpidly compare the input query gene against allergy and asthma biomarkers. To perform BLAST analysis, users need to provide the amino acid sequence of protein in FASTA format as input to search within BLASTdb, as shown in Fig. 2e where the amino acid sequence of a human Raf-1 proto-oncogene serine thre- onine kinase is given as input. The output of BLAST search will then appear with the best hit, the corre- sponding BLAST score and an E-value describing the significance of the search. The BLAST output page is Correspondence: Swati Gupta Bhattacharya, Division of Plant Biology (Main Campus), Bose Institute, 93/1 AcharyaPrafulla Chandra Road, Kolkata-700009 West Bengal, India. E-mail: swati@jcbose.ac.in and Sudipto Saha, Bio-informatics Center, Bose Institute (Centenary Building), P 1/12, C. I. T. Road, Scheme, VIIM, Kolkata, 700054 West Bengal, India. E-mails: ssaha4@gmail.com, ssaha4@jcbose.ac.in *Equal contribution.