GenesLove.Me: A Model-based Web-application for Direct-to-consumer

Genetic Tests

Jos

e Fabi

an Reyes Rom

1,2

, Carlos I

niguez-Jarr

ın

1,3

and

Oscar Pastor L

opez

Research Center on Software Production Methods (PROS), DSIC,

Universitat Polit

ecnica de Val

encia, Camino Vera s/n. 46022, Valencia, Spain

Department of Engineering Sciences, Universidad Central del Este (UCE),

Ave. Francisco Alberto Caama

no De

o, 21000, San Pedro de Macor

ıs, Dominican Republic

Departamento de Inform

atica y Ciencias de la Computaci

on,

Escuela Polit

ecnica Nacional, Ladr

on de Guevara E11-253, Quito, Ecuador

Keywords:

GLM, Direct-to-consumer Genetic Tests, Genetic Test, BPMN, Conceptual Models.

Abstract:

The objective of this work was to enhance personalized medicine through the development and implementation

of Genomic Information Systems (GeIS). For this, a web application called ”GenesLove.Me” (GLM) was

developed to provide direct-to-consumer genetic tests (DCGT). This paper focuses on the development of the

business processes (BPMN) and a conceptual model (CM) for GLM, designed to analyze and improve the

processes involved in this type of service and provide a model-based platform to manage genetic diagnoses

in a scalable, secure and reliable way. Software Engineering (SE) approaches applied to the genomic context

play a key role in the advancement of personalized and precision medicine.

1 INTRODUCTION

The current availability of direct-to-consumer genetic

tests (DCGT) has a great number of advantages for

the genomic domain, making it easier for end-users

to access early genetic-origin diseases diagnosis ser-

vices.

Romeo-Malanda (Romeo-Malanda, 2009) deﬁnes

“direct-to-consumer genetic analysis” as a term

which is used to describe analytic services offered to

detect polymorphism and health-related genetic vari-

ations. Although this type of analysis is available

through direct sales systems in pharmacies or other

health care bodies, the Internet has become the main

selling channel for direct-to-consumer genetic analy-

ses. The usual procedure is to take a biological sam-

ple at home and send it to an analysis laboratory. The

ﬁndings (results) of the analysis are communicated to

the client by telephone, mail or electronic mail, or

through secure access to an Internet portal (UNESCO,

2004).

The heterogeneity and dispersion of the data

sources represent a great challenge, known today as

”genomic chaos” (Le

on et al., 2016). As the man-

agement of genomic repositories offers many bene-

ﬁts to the biomedical community, it is necessary to

carry out studies and analyses to support the imple-

mentation of mechanisms to improve aspects related

to data integrity, consistency and homogeneity (Reyes

Rom

an, 2014). In this way it will be possible to gen-

erate more reliable results for end-users by guarantee-

ing true Precision Medicine (PM).

The genomic data repositories (e.g., NCBI (NCBI

Resource Coordinators, 2013), OMIM (Hamosh

et al., 2005) and Ensembl (Cunningham et al., 2015)

contain a very extensive set of information capable of

being analyzed to extract concise data and generate

results with greater accuracy. The efﬁcient use of ad-

vances in genomic research allows the patient to be

treated in a more direct way, which is reﬂected in re-

sults such as:”better health” and ”quality of life”.

In order to treat the data that will be used in

the proposed application, we implemented the SILE

(Search-Identiﬁcation-Load-Exploitation) methodol-

ogy (Reyes Rom

an and Pastor L

opez, 2016) which

was conceived in the PROS’s Research Center in or-

der to improve selective loading processes of our Hu-

man Genome Database (HGDB).

Our aim in this work was twofold:

• To provide “GenesLove.Me” (GLM) as a web ap-

plication designed to generate direct-to-consumer

genetic tests (DCGT) supported by Business

Román, J., Iñiguez-Jarrín, C. and López, Ó.

GenesLove.Me: A Model-based Web-application for Direct-to-consumer Genetic Tests.

DOI: 10.5220/0006340201330143

In Proceedings of the 12th International Conference on Evaluation of Novel Approaches to Software Engineering (ENASE 2017), pages 133-143

ISBN: 978-989-758-250-9

133

Process modeling (BPMN) (Object Management

Group, 1997) and conceptual modeling (MC)

(Oliv

e, 2007) techniques to study and analyze the

essential elements of the processes involved in the

genomic domain and improve the development of

Genomic Information Systems (GeIS).

• To apply Software Engineering (SE) knowledge

to improve Precision Medicine (PM). Ensuring

PM for end-users involves implementing tools for

studying, treating, exploring and exploding the

genomic data generated and stored for several

years to generate genetic diagnosis. The current

DCGT demand was the reason for studying and

analyzing the processes and stakeholders involved

in this domain. The genetic tests are the result of

the progress in the genomic environment with the

aim of exploiting PM.

The genomic domain requires methodologies and

modeling techniques capable of integrating innova-

tive ideas into data management, process improve-

ment and the inclusion of quality standards.

The paper is divided into the following sections:

Section 2 reviews the present state of the art. Section

3 describes BPMN applied to genetic testing. Section

4 contains the representation of the domain by a con-

ceptual model. Section 5 describes a case study with

the GLM application, and Section 6 summarizes the

lessons learned and outlines future work.

2 RELATED WORK

Bioinformatics now play an important role in con-

tributing advances to the medical and technological

sector. Genetic testing reveals existing knowledge

about ”genes” and ”variations” in the genomic do-

main, which is used to diagnose diseases of genetic

origin in order to prevent or treat them. This brings

precision medicine closer to end-users (i.e., clients /

patients).

The study of genomics (i.e., data repositories, ge-

netic variations, diseases, treatments, etc.) is con-

stantly growing and is increasingly seeking to ensure

the application of precision medicine. DNA sequenc-

ing began in 1977 and since then software tools have

been developed for its analysis. Thanks to NGS Tech-

nologies (Mardis, 2010), it is now possible to ma-

nipulate ﬁles (e.g., VCF) in order to generate genetic

diagnoses in a more agile and efﬁcient way (Reyes

Rom

an, 2014).

PM is a way of treating patients that allows doc-

tors to identify an illness and select the treatment most

likely to help the patient according to a genetic con-

cept of the disease in question (this is why it has also

been called Personalized Medicine) (Aguilar Carta-

gena, 2015).

Figure 1: From Genomics to Precision Medicine.

As shown in Figure 1, this approach is based on

a detailed knowledge of the genomic domain and on

the information derived from the large amount of data

generated in recent years. This information iscon-

stantly growing as research provides more and more

ﬁndings every day. Fowler et. al. (Jim

enez, 2014)

describe the advances in genomics that will provide

practical information on diseases, those who are most

likely to suffer them, and how most successful treat-

ments can be applied, thus reducing the uncertainty

and stress of patients and their families (Jim

enez,

2014), (Instituto Nacional del C

ancer, 2015).

The advantages of genetic tests are innumerable

and allow us to identify mutations or alterations in

the genes and are of great use and interest in clinical /

personalized medicine and the early diagnosis of dis-

eases (Fisiotraining C

ordoba, 2016), (Grupo RETO

Hermosillo, 2016). Bioinformatics supports the tasks

of: management of biological databases; metabolic

processes and population genetics; artiﬁcial intelli-

gence and others. The exploitation of genomics is

of interest to various branches of research, such as:

a) sequence analysis, b) genome annotation, c) evo-

lutionary computational biology, d) gene expression

analysis, e) protein expression analysis, f) cancer mu-

tation analysis, g) comparative genomics, h) model-

ing of biological systems, i) protein-protein coupling,

etc. (Franco et al., 2008)

By 2008 there were around 1,200 genetic tests

available around the world (Fisiotraining C

ordoba,

2016), (Grupo RETO Hermosillo, 2016), but they had

some limitations (e.g., data management, genome se-

quencing, etc.) and their cost was quite high. For this

reason, companies were interested in reducing costs

and providing services to end-users in the comfort of

their own homes. Technological advances played a

fundamental role in the genomic environment, since

the introduction of the NGS for sequencing samples

made it possible to obtain sequences more quickly

and cheaply (Metzker, 2010), (Voelkerding et al.,

2009).

23andMe (23andMe, 2016a), in this sector. This

American company offers a wide range of services.

The type of information obtained from genetic sam-

ples is oriented to: genetic history (ancestors) and

ENASE 2017 - 12th International Conference on Evaluation of Novel Approaches to Software Engineering

134

personal health (risk of diseases), and is presented

mostly in probabilistic terms (23andMe, 2016b). In

the same way, in Spain companies of this type have

emerged (e.g., Genotest

or IMEGEN

), all with the

aim of providing genetic tests to end-users, simply

and in the form of providing a diagnosis that allows

end-users to take reactive or corrective actions (e.g.,

prevention and treatment) to improve their quality of

life.

Genetic tests contain a large amount of sensitive

information, so that before offering these services it is

necessary to evaluate all the elements involved, such

as: ethical, moral, legal, etc. It is important to be

aware that the studies carried out must always become

prevention instruments oriented to helping end-users.

Companies within this business context treat sensitive

information, so it is necessary to implement security

(high quality) and data protection mechanisms, in ad-

dition to including all relevant legal measures. Span-

ish legislation requires these services to be provided

under the following laws:

1. Organic Law 15/1999, of December 13, on the

Protection of Personal Data (Consolidated text

dated March 5, 2011) (BOE, 1999).

2. Periods of Conservation of Personal Data in

Biomedical Research (BOE, 2011).

3. General Health Law 86 (BOE, 2015).

For the implementation of DCGTs it is necessary

to consider the aforementioned elements. In future

years we will continue to improve our understanding

of the human genome.

3 BPMN:

DIRECT-TO-CONSUMER (DTC)

GENETIC TEST

The GemBiosoft company is a spin-off of the Univer-

sitat Polit

ecnica de Val

encia (UPV), founded in 2010.

The main objective of this company is to de-

ﬁne the Conceptual Model of the Human Genome

(CMHG) (Reyes Rom

an et al., 2016), (Reyes Rom

et al., 2016) to obtain a precise schema to manage,

integrate and consolidate the large amount of ge-

nomic data in continuous growth within the genomic

domain. To achieve this objective, Gembiosoft has

extensive experience in Model Driven-development

(MDD) (Pastor et al., 2008) and an interdisciplinary

team of people -engineers and Ph.D.s- trained to im-

plement solutions aimed at i+D companies applying

http://www.trkgenetics.com/genotest

https://www.imegen.es/

information technologies in the bioinformatics and

health ﬁelds. GemBiosoft’s collaborators include:

PROS Research Center (UPV), PrincipiaTech and

IMEGEN (Instituo de Medicina Gen

omica).

3.1 Case Description

GemBiosoft has a web application called

”GenesLove.Me” which offers direct genetic

testing to the consumer. The information provided

by the genetic tests is accessible online to all users

without prior registration (anonymous users). Figure

2 depicts the general use case of the interaction

between actors and the web application, i.e. the ap-

plication’s functionality. For example, non-registered

users of the web application are able to consult all

information related to the diagnosis of rare diseases

of genetic origin, their characteristics, treatment,

tutorials and videos of the way in which the process

is performed.

Access security in GenesLove.Me is controlled by

proﬁles. Users can access GLM under 3 proﬁles: (1)

clients (patients), (2) provider and (3) administrator.

An authenticated user with a certain access proﬁle is

authorized to carry out the operations corresponding

to the access proﬁle.

(1) Clients (patients): Users with this proﬁle are

able to contract the services offered by selecting the

services (direct-to-customer genetic test) they are in-

terested in and then paying the fee. The user is able to

monitor the notiﬁcations and messages related to the

diagnoses, besides consulting the histories of all the

studies and treatments carried out and updating the

information associated with his proﬁle.

(2) Supplier: Users with this proﬁle are able to

generate notiﬁcations about the change of status in

the treatment of samples. After receiving the genetic

sample, the user activates the sample by entering its

code number. They can then track the sample until

the sequence ﬁle is generated. They can also update

their proﬁles and consult all the activated samples (in

progress and ﬁnalized).

(3) Administrator: A user with administration

privileges performs administration and maintenance

tasks of the web application, such as: a) publishing

online results (the administrator uploads the result-

ing diagnoses from the analysis performed on sam-

ples sequenced by the VarSearch

tool. The applica-

tion automatically notiﬁes the user when his/her re-

sults have been published); b) publishing advertise-

ments; c) publishing new diagnostic services to diag-

VarSearch is a web application that seeks variation (or

variant) ﬁles in a genomic database based on a Conceptual

Model of the Human Genome (CMHG).

GenesLove.Me: A Model-based Web-application for Direct-to-consumer Genetic Tests

135

Figure 2: General view of the system.

nose new diseases; d) consulting payment reports and

the application usage report (custom time period).

3.2 Genomic Diagnosis Process

Genetic tests are currently offered with the aim of de-

tecting a person’s predisposition to contracting a dis-

ease of hereditary origin (U. S. National Library of

Medicine, 2017). The bioinformatics domain seeks

to provide the necessary mechanisms and means to

generate genetic diagnoses that allow the end-users

(patients) to obtain these results to facilitate a person-

alized prevention treatment.

Figure 3 shows a BPMN diagram describing the

genetic diagnosis process (from the end-user’s ser-

vice request until he/she receives his/her genetic test

report). In this process, the three actors/users spec-

iﬁed in Section 3.1 are involved: 1) The client (pa-

tient) who requests the service to determine whether

or not he / she has a disease of genetic origin; 2) The

company, in this case GemBiosoft, which is in charge

of managing and performing the Genomic Diagnosis;

and 3) the Suppliers, who in this case prepare the ﬁle

containing the reference of the patient involved in the

genetic test.

The general process begins when the end-user (pa-

tient) enters the web application and requests the ge-

netic analysis (t1: task 1). The company (Gem-

Biosoft) processes this request and proceeds to send

the sample container to the client (t2). When the client

receives the container, he must activate it by regis-

tering its identiﬁer in the web application (t3), then

place the sample in the container and send it back to

the company (t4). Upon receipt of the sample, the

company conﬁrms that it meets the necessary require-

ments for the study and notiﬁes the customer of its

receipt (t5). The next step is to determine the sup-

plier who will be responsible for sequencing the sam-

ples and send him the sample (t6). The selected sup-

plier receives the sample and notiﬁes its reception to

the company (t7). Sequence preparation is initiated

through the sequencing technology used by the sup-

plier (t8). The supplier sends the resulting sequence

of the sample (ﬁle) to the company (t9). The company

conﬁrms its reception to the supplier and proceeds to

analyze the sequenced sample as part of the genetic

diagnosis (t10). The deﬁnitive diagnosis report (t11)

is then generated. The company (in this case the ad-

ENASE 2017 - 12th International Conference on Evaluation of Novel Approaches to Software Engineering

136

Figure 3: Genetic diagnosis process.

ministrator/user) proceeds to publish the genetic diag-

nosis (result) in the web application and the end-users

are automatically notiﬁed of the results (t12). To end

the process, the end-user accesses the web application

to obtain the diagnosis and make any queries (results)

(t13).

The BPMN (Business Process Model and Nota-

tion) gives companies the ability to understand their

internal business procedures in graphical notation and

the ability to communicate these procedures in a stan-

dard way (Chinosi and Trombetta, 2012). Through

the model shown in Figure 3, it facilitates the un-

derstanding of commercial collaboration and transac-

tions between organizations. In this ﬁgure we can see

the interactions between end-user, company and sup-

pliers (Reyes Rom

an, 2014). The company is inter-

ested in providing a web application that allows end-

users to obtain a quality genetic test in a simple way

that aids the treatment and prevention of diseases of

genetic origin.

4 REPRESENTATION OF THE

DOMAIN: CONCEPTUAL

MODEL

It is widely accepted that applying conceptual mod-

els facilitates the understanding of complex domains

(like genetics) (Reyes R et al., 2016). In our case

we used this approach to deﬁne a model represent-

ing the characteristics and the processes of direct-to-

consumer genetic tests (DCGT).

One of the leading beneﬁts of CM is that it accu-

rately represents the relevant concepts of the analyzed

domain (Reyes Rom

an et al., 2016). After perform-

ing an initial analysis of the problem domain, the next

step is to design a domain representation in the form

of a CM. Our CM (Figure 4) evolved with the new dis-

coveries made in the ﬁeld of genomics (Reyes Rom

et al., 2016) in order to improve data processing to en-

sure effective precision medicine (PM). We can thus

see how CM gives positive support to the knowledge

in which precision medicine plays a key role (Reyes

Rom

an, 2014). It is important to highlight that the ad-

vantage of CM for representing this domain is that it

eases the integration of new knowledge into the model

(Reyes Rom

an et al., 2016).

After an analysis of the requirements required for

this work, important decisions were taken to arrive at

an adequate representation of the basic and essential

concepts in the understanding of the domain under

study. Figure 4 presents the conceptual model pro-

posed, which can be classiﬁed into three main parts:

a) Stakeholders, b) Genetic diagnostics and c) Sales

management.

The ﬁrst part of our conceptual model represents

all the participants involved in the web application.

To represent the ”Users” class of the inherit type we

have the ”Administrator”, ”Supplier Seq”, ”Patient

Client” and ”Anonymous” classes respectively

The DCGTs are initialized when the end-user

(”patient customer”) accesses the application. If the

interaction does not have user credentials, it remains

an ”anonymous” user, otherwise, it becomes a client.

The ”client customer” contracts one or more services,

which cover one or more diseases, represented in the

model by the ”Services” and ”Diseases” classes, re-

spectively.

Diseases are studied using the SILE methodology

(Reyes Rom

an and Pastor L

opez, 2016), and when

complete, the study is offered as a service. The end-

user adds the services of interest to his shopping cart

(represented by the ”Shopping Cart” class) and then

GenesLove.Me: A Model-based Web-application for Direct-to-consumer Genetic Tests

137

Figure 4: GLM conceptual model.

performs the payment process. For this, all the billing

information, shipping address and acceptance of the

purchase contract (which explains the rights and du-

ties of the end-user and the company) is entered in the

CM through the ”Payments” class.

The management of genetic diagnoses is repre-

sented in the conceptual model through the ”Diagnos-

tic” and ”Historical diagnosis” classes. The ”Diag-

nosis” class shows the results obtained after the anal-

ysis of the sequence and a list of the diseases con-

tracted. The user has the ability to consult his history

associated with all the diagnoses requested by the ap-

plication.

The diagnosis of genetic data begins when the

company receives the ﬁle containing the sample

sequenced by the supplier (represented by ”Sup-

plier Seq” class). The application administrator is re-

sponsible for publishing the diagnosis results on the

web and automatically notiﬁes the end-user of the

availability of his report. Other activities of the ad-

ministration consist of the publication of advertising

related to the implementation of new services and

control of sales of the application.

Through our CM we incorporate genetic data cur-

rently used in the PM, achieving a conceptual repre-

sentation that meets the needs of the bioinformatics

domain. As we mentioned above, this model aims to

improve the conceptual deﬁnition of the treatment re-

lated to genomic diagnosis, and thus leave a concep-

tual framework for further improvements.

5 CASE STUDY: GenesLove.Me

SERVICE FOR END-USERS

In this section, we describe the design and implemen-

tation of GenesLove.Me

(GLM), an online web ap-

plication based on the business process described in

Section 3 of this work. The application becomes the

point of interaction between three actors: the clients

of genetic tests, the company that performs the genetic

diagnosis and the suppliers who sequence the genetic

samples. The main objective is to operate the DCGT

service through an accessible online and easy-to-use

platform for the actors involved in the process.

The platform offers a wide range of genetic tests

on its main page (Figure 5), from which users are

able to contract one or several genetic testing services

(e.g., lactose intolerance, androgenetic alopecia, al-

cohol sensitivity, etc.) according to their needs. The

company in charge of genetic analysis and diagno-

sis manages all the service orders received and mon-

itors the change of order status throughout the ser-

vice delivery process. For instance, the status of the

order changes to ”KIT SENT TO SUPPLIER” when

the company (in charge of the analysis and diagno-

http://www.geneslove.me

ENASE 2017 - 12th International Conference on Evaluation of Novel Approaches to Software Engineering

138

Figure 5: GenesLove.Me’s home web page showing the available direct-to-customer genetic tests.

sis) sends the genetic sample to the suppliers respon-

sible for sequencing the genetic samples and once the

genetic samples have been sequenced and received

from the supplier, the order is assigned ”ANALYSING

SAMPLE” status. Each status change is notiﬁed to

the end-user via email and can be further visualized

on the web user interface.

GLM is a web application implemented under a

client-server architecture as is shown in Figure 6. The

client side, the user’s browser (e.g., Firefox (Mozilla,

2017), Chrome (Google, 2017), Internet Explorer

(Microsoft, 2017), etc.) serves as the interaction point

between the user and application. The end-user in-

teracts with an easy-to-use graphical web interface to

request genetic testing services available in the ap-

plication. The server side is hosted on the Internet

and contains: a) the Apache 2.2 web server (The

Apache Software Foundation, 2017) with the applica-

tion logic implemented with PHP (W3schools, 2017)

programming language and b) the data stored in the

MySQL5.5 (Oracle Corporation, 2017) database en-

gine.

The GLM design allows customers to access the

online product range (i.e., clinical tests) from any-

where and at any time. In the same way, the adminis-

trator users in charge of site management and internal

business tasks can access the GLM through an Inter-

net connection.

Database

(MySQL)

Web Server (Apache)

Server

http

internet

browser’s

user

Geneslove.me

Application

(PHP)

Figure 6: GenesLove.Me architecture.

GenesLove.Me: A Model-based Web-application for Direct-to-consumer Genetic Tests

139

GLM is implemented under Prestashop

, an open

source eCommerce CMS platform, which facilitates

the implementation of customized solutions oriented

to the marketing of products framed in a simple

purchase-sale process. The platform incorporates

both modules and website templates in order to pro-

vide, respectively, speciﬁc functionality and graphic

style customized according to the business’s needs.

There is a great variety of free and commercial mod-

ules

and website templates created by web develop-

ers available on the Internet to be down- loaded and

used.

The default Prestashop download package avail-

able on the ofﬁcial site includes modules of basic

functionality (e.g., customers, products, orders, etc.)

which are sufﬁcient to create and manage a basic

e-commerce platform. However, Prestashop allows

complex functionality modules to be incorporated to

tailor sites to particular needs. For instance, it is pos-

sible to install a module that deals with the complexity

associated with the credit card payment process.

GLM takes full advantage of the range of func-

tionalities offered by a CMS e-commerce platform

such as Prestashop. Indeed, GLM offers a variety

of on- line genetic tests, keeps the customer database

and orders the genetic test sales process thanks to the

available payment methods, such as bank transfer,

credit card payments through PayPal (Jackson, 2012)

or payments by electronic check.

The GLM client module represents the database of

registered clients interested in the genetic tests. The

ordering module lists the requests made by the cus-

tomers, which consists of a reference code useful for

order tracking, the name of the customer, the total

value, the payment method and the current status of

the order within the business process. The applica-

tion’s administrator manages the orders and updates

the state of the orders. Updating the order state means

a status change that automatically notiﬁes the end user

via e-mail. GLM facilitates the management of orders

through the sequence of 13 states shown in Figure 7.

Every state order change generates an email notiﬁca-

tion to keep customers informed of the processing of

their orders. It is important to note that GLM includes

interaction with VarSearch (see task 10 and 11 of Fig-

ure 3), an application developed by PROS Research

Center to automatically identify the relevant informa-

tion contained in the genomic databases and directly

related to the genetic variations of the sequenced sam-

ple. VarSearch relies heavily on a Conceptual Model

of the Human Genome (CMHG), which makes inte-

gration of external genomic databases feasible. How-

https://www.prestashop.com

(http://addons.prestashop.com/en/2modules)

Figure 7: GLM’s states sequence to manage orders.

ever, due to the large amount of information available,

the data loaded in VarSearch are the result of a selec-

tive loading process (Reyes Rom

an and Pastor L

opez,

2016) where the selected data correspond to the rele-

vant information on the disease to be analyzed.

ENASE 2017 - 12th International Conference on Evaluation of Novel Approaches to Software Engineering

140

The GLM user administrator has access to the ad-

ministration panel (back-ofﬁce) to manage the entire

business process as well as the security and function-

ality of the site. The management screen has relevant

information and direct access to the conﬁguration and

maintenance zones. For example, a strategically vis-

ible part of the main management screen displays in-

formation on the sales indicators and useful informa-

tion for the management of the business accounting.

By accessing the access security module, it is possi-

ble to conﬁgure proﬁles and access permissions for

all users. It is also possible to manage the modules

of users, products, orders, as well as the rules for dis-

counts on purchases and customer service (e.g., fo-

rums, e-mail notiﬁcations). Thanks to the web plat-

form on which the solution is implemented, site man-

agement can be carried out from anywhere and at any

time.

Each module installed adds a functionality that

supports to a greater or lesser degree the main rea-

son of the platform: buying and selling products.

The buying-and-selling process establishes the buyer-

seller relationship through a simple process consisting

of three states: Pending, Paid and Sent. ”Pending” in-

dicates that the user has requested a product, stored

it in the shopping basket but still has to generate the

payment. The ”Paid” status indicates that the user

has registered the payment for any of the options con-

ﬁgured and available (e.g., bank transfer) and ”Sent”

indicates that the seller has shipped the product and is

in the process of delivery to the customer.

5.1 Process Validation

In order to validate the process proposed in Section 3,

test cases were performed with the implemented solu-

tion. The validation scenario consisted of a group of

ﬁve (5) users, who made requests for genetic testing

for ”lactose intolerance”. To begin the process, each

user involved in the case study authorized the proce-

dure through an ”informed consent” (Reyes Rom

an,

2014), (de Galicia, 2001) which becomes a legal sup-

port that establishes the rights and obligations of the

service offered and its expected scope.

6 LESSONS LEARNED AND

FUTURE WORK

This paper describes a study and analysis of the

implementation of a web application to facilitate

DCGT, which informs end-users on their predispo-

sition to suffer certain genetically based illnesses.

Through the development of our web application

”GenesLove.Me” we seek to provide end-users with

a genomic diagnosis in a secure and reliable way.

The use of BPMN and Conceptual Modeling-

based approaches for this type of service aids the

understanding of the participants in the processes in

the genomic domain and improves the processes in-

volved. Through the proposed models we were able

to evaluate different points:

1. Process evaluation: the processes involved in

the direct-to-consumer genetic tests (DCGT) were

studied, and a business model was developed to

deﬁne the structures and relationships with the

stakeholders.

2. Improvement in data management: a conceptual

model (CM) was proposed for the deﬁnition of

the relevant concepts in the domain with the ob-

jective of guaranteeing reliable and personalized

medicine.

The validation of -GenesLove.Me- included the

study of four disorders of genetic origin: Alcohol

Sensitivity, Androgenic Alopecia, Lactose Intoler-

ance, and Dupuytren. The research process was car-

ried out using the SILE (Search-Identiﬁcation-Load-

Exploitation) Methodology, and after obtaining the

variations/genes associated with the diagnoses, we

made the selective loading of the data to our HGDB

(Human Genome Database).

This web application was developed to provide

end-users with an early genetic diagnosis through an

attractive and easy-to-use distribution channel. Bioin-

formatics is a domain that is constantly evolving, and

with the application of conceptual models we can ex-

tend our genomic knowledge and conceptual repre-

sentation accurately and simply.

Future research work will focus on three main

goals:

• The study and treatment of new diseases of ge-

netic origin and continue expanding the list of ill-

nesses available in the web application.

• To extend our Conceptual Model of the Human

Genome (CMHG) by integrating new genomic in-

formation , to improve the generation of genetic

diagnoses.

• Implementation of data management mechanisms

to enhance the quality of personalized medicine.

ACKNOWLEDGEMENTS

The author thanks to the members of the PROS Cen-

ter’s Genome group for fruitful discussions. This

work was supported by the Ministry of Higher

GenesLove.Me: A Model-based Web-application for Direct-to-consumer Genetic Tests

141

Education, Science and Technology (MESCyT) of

Santo Domingo, Dominican Republic and the Sec-

retar

ıa Nacional de Educaci

on, Ciencia y Tecnolog

ıa

(SENESCYT) and the Escuela Polit

ecnica Nacional

de Ecuador. The project also had the support of

the Generalitat Valenciana through Project IDEO

(PROM- ETEOII/2014/039) and the Spanish Ministry

of Science and Innovation through Project DataME

(ref: TIN2016-80811-P).

The author thanks Alberto Garc

ıa S., Mercedes

R. Fern

andez Alcal

a, Vicente Mart

ınez Perell

o and

Ver

onica Burriel Coll for their collaboration with this

this project.

REFERENCES

23andMe (2016a). 23andMe. https://www.23andme.com/.

[Online; accessed 19-December-2016].

23andMe (2016b). How it works.

https://www.23andme.com/howitworks/. [Online;

accessed 19-December-2016].

Aguilar Cartagena, A. (2015). MEDICINA PERSON-

ALIZADA, MEDICINA DE PRECISI

ON, ¿CU

LEJOS ESTAMOS DE LA PERFECCI

ON? Carci-

nos, 5:1–2.

BOE, A. e. B. O. d. E. (1999). Disposiciones Generales.

BOE, A. e. B. O. d. E. (2011). Ley 14/2007 de Investigaci

Biom

edica.

BOE, A. e. B. O. d. E. (2015). Ley 14/1986 General de

Sanidad.

Chinosi, M. and Trombetta, A. (2012). BPMN: An intro-

duction to the standard. Computer Standards and In-

terfaces, 34(1):124–134.

Cunningham, F., Amode, M. R., Barrell, D., Beal, K., Billis,

K., Brent, S., Carvalho-Silva, D., Clapham, P., Coates,

G., Fitzgerald, S., Gil, L., Gir

on, C. G., Gordon, L.,

Hourlier, T., Hunt, S. E., Janacek, S. H., Johnson,

N., Juettemann, T., K

ari, A. K., Keenan, S., Mar-

tin, F. J., Maurel, T., McLaren, W., Murphy, D. N.,

Nag, R., Overduin, B., Parker, A., Patricio, M., Perry,

E., Pignatelli, M., Riat, H. S., Sheppard, D., Taylor,

K., Thormann, A., Vullo, A., Wilder, S. P., Zadissa,

A., Aken, B. L., Birney, E., Harrow, J., Kinsella, R.,

Muffato, M., Rufﬁer, M., Searle, S. M. J., Spudich, G.,

Trevanion, S. J., Yates, A., Zerbino, D. R., and Flicek,

P. (2015). Ensembl 2015. Nucleic Acids Research,

43(D1):D662–D669.

de Galicia, C. A. (2001). Ley 3/2001, reguladora del con-

sentimiento informado y de la historia cl

ınica de los

pacientes.

Fisiotraining C

ordoba (2016). Test Gen

eticos.

http://www.ﬁsiotraining.com/cordoba/

analisisgeneticos

quees.htm. [Online; accessed

19-December-2016].

Franco, M. L., Cediel, J. F., and Pay

an, C. (2008). Breve

historia de la bioinform

atica. Colombia M

edica,

39:117–120.

Google (2017). Personalizar Chrome.

Grupo RETO Hermosillo, A. (2016). El c

ancer de mama.

http://gruporetohermosilloac.com/index.php. [Online;

accessed 19-December-2016].

Hamosh, A., Scott, A. F., Amberger, J. S., Bocchini, C. A.,

and McKusick, V. A. (2005). Online Mendelian Inher-

itance in Man (OMIM), a knowledgebase of human

genes and genetic disorders. Nucleic Acids Research,

33(DATABASE ISS.):D514–D517.

Instituto Nacional del C

ancer (2015). Qu

e es la

medicina de precisi

on en el tratamiento del c

ancer.

https://www.cancer.gov/espanol/cancer/tratamiento/

tipos/medicina-de-precision. [Online; accessed 19-

December-2016].

Jackson, E. M. (2012). The PayPal Wars: Battles with eBay,

the Media, the Maﬁa, and the Rest of Planet Earth.

WND Books, 2nd edition.

Jim

enez, N. (2014). Una medicina nueva , m

as inteligente

y menos invasiva. http://www.lifescienceslab.com.

on, A., Reyes, J., Burriel, V., and Valverde, F. (2016).

Data quality problems when integrating genomic in-

formation. In International Conference on Conceptual

Modeling, pages 173–182. Springer.

Mardis, E. R. (2010). The $1,000 genome, the $100,000

analysis? Genome Medicine, 2(11):84.

Metzker, M. L. (2010). Sequencing technologies - the next

generation. Nature reviews. Genetics, 11(1):31–46.

Microsoft (2017). Internet Explorer - Microsoft Download

Center.

Mozilla (2017). Firefox.

NCBI Resource Coordinators (2013). Database resources

of the National Center for Biotechnology Information.

Nucleic Acids Research, 41(D1):D8–D20.

Object Management Group (1997). Business Process

Model and Notation.

Oliv

e, A. (2007). Conceptual Modeling of Information Sys-

tems. Springer Berlin Heidelberg, Berlin, Heidelberg,

1 edition.

Oracle Corporation (2017). MySQL :: MySQL 5.7 Refer-

ence Manual.

Pastor, O., Espa

na, S., Panach, J. I., and Aquino, N.

(2008). Model-driven development. Informatik-

Spektrum, 31(5):394–407.

Reyes R, J. F., Pastor, O., Valverde, F., and Rold

an, D.

(2016). Including haplotypes treatment in a Ge-

nomic Information Systems Management. In Pro-

ceedings of the XIX Ibero-American Conference on

Software Engineering (CIbSE 2016), pages 27–29,

Quito, Ecuador.

Reyes Rom

an, J. F. (2014). Integraci

on de haplotipos al

modelo conceptual del genoma humano utilizando la

metodolog

ıa sile. http://hdl.handle.net/10251/43776.

Reyes Rom

an, J. F., Pastor,

O., Casamayor, J. C., and

Valverde, F. (2016). Applying conceptual modeling

to better understand the human genome. In Comyn-

Wattiau, I., Tanaka, K., Song, I.-Y., Yamamoto, S.,

and Saeki, M., editors, Conceptual Modeling: 35th

International Conference, ER 2016, Gifu, Japan,

November 14-17, 2016, Proceedings, pages 404–412.

Springer International Publishing, Cham.

ENASE 2017 - 12th International Conference on Evaluation of Novel Approaches to Software Engineering

142

Reyes Rom

an, J. F., Pastor Lopez, O., Valverde, F., and

Rold

an, D. (2016). How to deal with Haplotype data :

An Extension to the Conceptual Schema of the Human

Genome. CLEI ELECTRONIC JOURNAL, 19(2).

Reyes Rom

an, J. F. and Pastor L

opez,

O. P. (2016). Use

of GeIS for Early Diagnosis of Alcohol Sensitivity.

In Proceedings of the 9th International Joint Confer-

ence on Biomedical Engineering Systems and Tech-

nologies, pages 284–289.

Romeo-Malanda, S. (2009). An

alisis gen

eticos direc-

tos al consumidor: cuestiones

eticas y jur

ıdicas.

http://www.institutoroche.es/legalactualidad/85/

analisis. [Online; accessed 19-December-2016].

The Apache Software Foundation (2017). Welcome! - The

Apache HTTP Server Project.

U. S. National Library of Medicine (2017). What is genetic

testing? - Genetics Home Reference.

UNESCO (2004). Declaraci

on Internacional

sobre los Datos Gen

eticos Humanos.

http://unesdoc.unesco.org/images/0013/001361/

136112so.pdf. [Online; accessed 19-December-

2016].

Voelkerding, K. V., Dames, S. A., and Durtschi, J. D.

(2009). Next-generation sequencing:from basic re-

search to diagnostics. Clinical Chemistry, 55(4):641–

658.

W3schools (2017). PHP 5 Tutorial.

GenesLove.Me: A Model-based Web-application for Direct-to-consumer Genetic Tests

143