1 Introduction

Recent high-throughput technological advancements in next generation sequencing and microarray have generated huge data at relatively lower cost (Diamandis, 2009; Fehniger and Marko-Varga, 2011; Glenn, 2011; Merdad et al., 2014). Biobanks are established for long-term storage and conservation facility for biological specimens along with their demographic, clinical and experimental information to support scientific investigation using bioinformatics tools (Artene et al., 2013). Once the samples size grow in thousands then manual methods fail in eficient handling and samples information can also be lost. Quality becomes another big concern with big data. Therefore, need of a robust software arises to manage biobank and laboratory information for assuring quality of results with approved ethical guideline and personal integrity ((ISBER) 2012; Bredenoord et al., 2011; Kang et al., 2013; Voegele et al., 2007). Softwares are helpful in managing the data flow cycle comprising of collection, storage, analysis and report generation to facilitate quick and easy retrieval of information and speed up biomarker and therapeutic discoveries (Melo et al., 2010).

Collection of significant cohort size with crucial factors like number and type of samples, clinical information, pathological finding, follow-up data etc. is a time taking process but strongly recommended to facilitate comprehensive translational research (Betsou et al., 2010; Hewitt, 2011; Huang et al., 2011; Riegman et al., 2008). Bioinformatics softwares are either available as commercial with proprietary code and high cost or academic/open source with complexity that are difficult to adopt with other laboratories (Greely, 2007; Huang, Arkin, and Chandonia, 2011; Kauffmann and Cambon-Thomsen, 2008; Minamikumo, 2012; Prilusky et al., 2005). We, therefore, developed and implemented a software to support the investigators to access all clinical and experimental disease associated information required for basic and translational research. Herein, we discuss the application of our in-house developed software for genomic analysis specifically for breast cancer transcriptomics and can be used further for any diseases.

2 Materials and methods

Software was developed using three-tier architecture model: (i) Presentation tier – an interactive web browser for end user’s computer, (ii) Data tier – a SQL Server Management Studio that manages the storage and DB-server, and (iii) Business tier – acting as bridge between the rest two and collects data from the presentation tier, checks for validations and finally sends them to the data tier and vice versa. The BS follows standard guidelines like legislation agreement, standardized architecture, workflow and dedicated staff members, and technical procedures to record and access clinical data as described below:

3 Results

In the past thirteen years (2007–2020) we collected significantly high number of specimens (n = 25,396) and their derivatives (n = 27,882) with number growing every year (Fig. 1, Table 1). Therefore, we established a disease-oriented biobank and developed Biosearch System (BS), an in-house database software to manage biobank and store disease associated genomic information.

Close

Fig. 1

Collection and storage of specimen and disease associated information at CEGMR/KAU biobank. Pictorial graph showing progressive growth of specimen/derivatives from zero in 2007 to twenty-seven thousand till 2020.

Export to PPT

A prototype of an entity-relationship diagram was developed to depict numerical relationship among different entities like sample, storage condition, specimen request, approval system, project type, diagnosis, and disease type (Fig. 2). We established a communication between BS host server using SQL Server Management Studio and employed ASP.NET and JavaScript/Ajax toolkit for end users (Fig. 3). It was divided into 3 main parts: (i) data and specimen acquisition, (ii) data management and (iii) specimen distribution/request (Fig. 4). It classifies the user in four groups (biobank technicians, researchers, lab manager and director) and hierarchal permission is given accordingly. Administrator of BS has power to do all function, including addition of new users and granting permission according to hierarchy (Table 2).

Close

Fig. 2

Entity-relationship diagram. An entity-relationship diagram for database structure used for Biosearch System depicting numerical relationship among different entities like sample, storage condition, specimen request, approval system, project type, diagnosis, and disease type.

Export to PPT

Close

Fig. 3

Black box of system structure. The relationship and the communication process between server side and client side; where Ajax and javascript sends client request to the SQL server side and ASP.net shows the processed response back at client end using the graphical user interface.

Export to PPT

Close

Fig. 4

Biological sample life cycle. Sequential flow of each biobank specimen starting from collection, processing of sample, storage, maintenance, request, retrieval, distribution, and utilization.

Export to PPT

BS is supported by a data warehouse and query tools interface that can be used efficiently to search and short list patients as per a particular criterion without disclosing their privacy. Based on evaluation form analysis we found that it is robust, secure, flexible, efficient, and user-friendly database. Robust, as it is platform independent, compatible with any version of window. Scripts are free from error, so no hanging of system. Secure, as in addition to appropriate permission strategy for different users and password protection, KAU network security strategy was also incorporated to transfer data safely, where each user has his own id and password to access specific computer and the password is stored in database in encrypted way. Each machine that has access to the BS must has its own firewall and anti-virus Trend Micro office scan agent provided by KAU IT deanship.

The system takes backup at regular interval (monthly backup) on KAU servers, so any loss of data from the system can be easily restored thereby saving the date from any disasters. Flexible, as new features/modules can easily be added by authorized users to customize according to laboratory and researcher requirements. Efficient, as it has been managing huge amount of clinical and experimental data of CEGMR competently since 2014. User-friendly, as simple icons, buttons, drop-down lists etc. are convenient for users and they can easily add or retrieve data using simple icons with available multiple options.

Our performance and utility of software evaluation with the help of real time feedback from users was excellent with satisfaction score was 4.7/5. Results of performance test of search, insert, update, and delete function are also satisfactory (Table 3). BS is efficient in term of number of transactions and number of users to maintain the quality of data and permits pliability in the workflow.

BS defers from other LIMS based on its customization approach for CEGMR laboratory and is flexible enough to be adjusted for any changes in future to suit the needs of projects and researcher such as new equipment, procedure, and/or software to complement or improve the workflow without changing the core code. For example: essential customization needed for research purpose is to link SMARTGENE software, a system for diagnostic lab, with BS system and exchange data with researchers consent. It can be used for any laboratory with minor modifications (QC, R&D, and analytical service) whereas most of the commercial software are a bit rigid. Its flexibility extends to data quality by achieving through the computerization (calculations, statistics) and automation of processes, and procedures to minimizing manual lab tasks such as data entry. For instance, the system makes sure entering data occurs in a standardized way using drop down pre-defined list for most processes. Presently BS is not available as open source because of institutional policy but we encourage the interested researchers to request for codes on individual basis.

BS is efficient to maintain the quality of data and permits pliability in the workflow and it had synchronized the biobank system of CEGMR/KAU. Current state of the CEGMR biobank is as follows: (1) Infrastructures: biobank is well equipped with liquid nitrogen cylinder (−196 °C), deep freezer (−80 °C and −20 °C), refrigerator and UPS system. (2) Personnel: presently fourteen dedicated staff are working hard for smooth running of biobank unit, and (3) Patient’s samples management system is fully functional and regularly handling the dispatch, collection, processing, and storage of thousands of specimens. The BS can be accessed by authorized person using the login username only (Fig. 5). Our clinical database contains the following clinicopathological parameters: CEGMR code for patient, receiving date, hospital MRN number, name, date of birth, age, sex, nationality, disease, date of diagnosis, status, filing date, histology, sites, grade, size, lymph node status, invasion, margin status, immunohistochemical data, family history, medication, follow up etc. Similarly, biobank management system contains specimen and their derivative related information: type of specimen, receiving date, extraction date, storage of specimen and its derivatives, quality and quantity record, handling request of researchers, distribution, and maintenance of specimen.

Close

Fig. 5

A glimpse of Biosearch System. Screenshot of the new sample addition page of BS allowing authorized researchers to access samples and their disease associated genomic and clinical information.

Export to PPT

3.1

3.1 Identification of DEGs for breast cancer

We conducted genome wide expression study to understand the molecular phenomenon leading to triple negative breast cancer and found 1181 differentially expressed genes with P-value <0.05 and FC 2. The upregulated genes for TNBC were IFI6, LEF1, CCR8, FANCI, TRIM59, CASC5, and PLXNA3 while down regulated genes were ADH1B, LYVE1, ADH1C, ADH1B, FIGF, ADIPOQ, PLIN1, and LYVE1. Hierarchical clustering of top 135 DEGs showing distinct pattern for genes in triple negative BC and control samples (Fig. 6).

Close

Fig. 6

Hierarchical clustering of DEGs. Unsupervised clustering showing expression pattern of genes in triple negative BC. Blue and red colors indicating down and up-regulated genes. Row and Column represents samples and DEGs respectively.

Export to PPT

3.2

3.2 Pathways associated with triple negative breast cancer (TNBC)

Molecular pathway analysis revealed more than hundred canonical pathways where most activated pathways were (i) interferon signaling pathways (z-score = 2.646) with participating genes IFI6, FNG, IRF1, IRF9, MED14, MX1, PSMB8, STAT1, STAT2, and (ii) kinetochore metaphase signaling pathway (z-score = 2.138) with following associated genes; BUB1, BUB1B, CDK1, CENPL, KIF2C, KNL1, KNTC1, NUF2, PLK1, PPP1CA, PPP1R14B, PTTG1, RAD21, REC8 and TTK (Fig. 7).

Close

Fig. 7

Interferon Signaling. Overall, the pathway is predicted to be upregulated and participating differentially expressed genes are IFI6, FNG, IRF1, IRF9, MED14, MX1, PSMB8, STAT1 and STAT2.

Export to PPT

4 Discussion

Today’s biobanks are much more than just sample repository. They store a huge amount of clinical and experimental data related to specimen (Calleros et al., 2012). However, efficient data management is a bottleneck in the genomic medicine research process. It has been observed on many occasions that researchers face difficulties in data collection, maintenance, follow up studies and sometime publish the work with missing data (Greely, 2007; Kauffmann and Cambon-Thomsen, 2008; Minamikumo, 2012). Translational genomic research is based on a secure database and biobank system. A well-designed software provides the possibility of finding significant associations among stored information and facilitates diagnostics and therapeutics research (Lemmon et al., 2011; Zerhouni, 2005). BS is used to collect, store, distribute and maintain the specimen data, as well as its relevant clinical and experimental information to support ongoing research that can lead to discovery of novel cancer biomarkers and therapeutics targets (Karim et al., 2016; Karim et al., 2019; Merdad et al., 2014; Merdad et al., 2015; Mirza et al., 2015; Mirza et al., 2014; Rasool et al., 2021; Rasool et al., 2020; Schulten et al., 2016; Subhi et al., 2020; Sultan et al., 2021).

We used SQL server management studio, Visual Basic, .NET and Ajax control kit to develop interactive database, and ASP.NET for better code management, clean code structure and fast web applications. The first prototype of BS was ready in 2010 and after feasibility test and successful trial run, we released the present final version with minor modifications. BS is web-based tool accessible from off-and-on-campus. High level of protection allows only authorized researchers to access database. The security concerns like threat of hacking, virus attack etc were taken care of without any compromise (Bjugn and Hansen, 2013; Mintzer et al., 2013; Rogers et al., 2011; Simeon-Dubach et al., 2013). Bioethical safety was another big concern and needed proper care in establishing biobank and developing database ((ISBER) 2012; Bredenoord et al., 2011; Hansson 2009; McGuire et al., 2008; Rotimi and Marshall, 2010).

BS makes the job easier for everybody involved in research: (i) clinician and pathologist can provide clinicopathological information of each provided specimen from hospitals/clinics through online database system, (ii) biobank staff can store and maintain the specimen and extracted derivatives easily, and (iii) researchers can extract clinical information and request specific specimen/derivative as per requirement of on going project (Kang et al., 2013). Presently our data set is small, and we are collecting selected cases only, so the frequency of disease should not be used to represent society at large. However, the frequency of different type of cancer in our biobank is very much like national cancer registry. In future, this can pave the way for expanding this database in association with Saudi cancer registry or any other national level databases.

BS tool was tested by CEGMR staffs, and was found satisfactory. The users were comfortable to organize all sample related information and found it user friendly with simple icons, buttons, drop-down list etc. To ensure the safety of data, King Abdulaziz University network security was utilized and only user with approved permission and approved supporting with anti-virus were allowed to access. Periodic system backup strategy also ensured avoidance of any data loss due to unseen disaster. It is customized according to actual activities and workflow in CEGMR lab and is flexible enough to adopt new modules to add more features in future.

5 Conclusions

Biosearch system is a user’s friendly software to manage biobank specimen with clinical information to facilitate genomic medicine research leading to discovery of disease biomarkers and therapeutic targets. It is adoptable to new features and modules to add barcoding system, quality control system and reagent purchasing system in future.

CRediT authorship contribution statement

Sajjad Karim: Conceptualization, Data curation, Fund acquisition, Investigation, Project Administration, Software, Supervision, Writing - original draft. Adnan Merdad: Conceptualization, Fund acquisition, Validation, Writing - review & editing. Saddig Jastaniah: Conceptualization, Fund acquisition, Validation, Writing - review & editing. Sudhir Kumar: Conceptualization, Fund acquisition, Supervision. Adel Abuzenadah: Conceptualization, Project Administration, Resources, Validation, Writing - review & editing. Mohammed Al-Qahtani: Conceptualization, Project Administration, Resources, Validation, Writing - review & editing. Mona Alkharaz: Data curation, Formal analysis, Investigation, Methodology. Zeenat Mirza: Data curation, Investigation, Methodology, Software, Visualization, Writing - original draft. Mahmood Rasool: Data curation, Investigation, Validation, Writing - review & editing. Hend Noureldin: Formal analysis. Heba Abusamra: Formal analysis. Nofe Alganmi: Methodology, Resources, Software, Supervision, Visualization, Writing - review & editing.