Laboratory Information Management System for NGS Genomics Labs

Jitao Yang

1,2

, Shuai Zhu

, Guangpeng Xu

and Chao Han

School of Information Science, Beijing Language and Culture University, Bejing 100083, China

Annoroad Gene Technology, Beijing 100176, China

Keywords:

LIMS, NGS, Genome Sequencing, Genomics, Sample Tracking.

Abstract:

The goal of genome sequencing is to unravel the ordered sequence of nucleic acids that form the DNA or RNA

of a given sample. Genome sequencing lab requires the ability to select and track a large amount of samples

through many experimental steps. Therefore, laboratory information management system (LIMS) is needed

to provide a way of automating the laboratory experimental procedures and track the samples. LIMSs have

been proposed and developed for many years, but still remain difﬁcult for labs to implement successfully. In

this paper, we demonstrate our genomic next generation sequencing (NGS) LIMS solution. We developed a

web-based LIMS with ﬂexible conﬁguration and customization for NGS laboratories, and can help laboratories

track samples and optimize experimental procedures and business workﬂows. We also describe our solution of

integrating LIMS with the existing enterprise business information systems. Finally, we share our experience

for the implementation of a successful LIMS.

1 INTRODUCTION

Genome next generation sequencing is now com-

monly adopted, and has a broad areas of applica-

tions, such as the Non-invasive Prenatal DNA Testing,

ctDNA Testing for Non-invasive Tumor Personalized

Therapy, Plant and Animal Molecular Breeding, Ge-

netics and Evolution, Microorganism and Ecological

Environment, etc.

To manage tens of thousands of samples that are

subject to NGS analysis, it is inevitable to develop ad-

equate laboratory information management system to

track and manage the NGS workﬂows. However, it is

extremely difﬁcult to manage the NGS expriment and

analysis procedures accurately and efﬁciently, given

the issues of multiple-source samples enrolling, na-

tional and international logistical management and

tracking, fragmented procedures for assessment and

processing of samples, the intricacies of molecular

experimental steps, and the complex and multiple

pipelines of NGS processing.

LIMS provides many beneﬁts for the users of lab-

oratory, several of the main beneﬁts identiﬁed are out-

lined below:

• brings accuracy and accessibility to the ﬂow of

samples and data in laboratory,

• universally accessible data via the web rather than

digging through ﬁles,

• years of data can be kept and queried conve-

niently,

• business efﬁciency improvement,

• data quality control,

• efﬁcient sample tracking and management,

• automated and in-depth customer reports,

• integration with laboratory instruments,

• experimental steps quality control,

• building automated analysis pipelines,

• status and results sharing with collaborators and

clinicians,

• ﬁnancial management,

• access control,

• track and analyze trends,

• error reduction.

In this paper, we demonstrate our genomic NGS

LIMS solution. Our LIMS is a web-based system

with ﬂexible conﬁgurations and customizations for

NGS laboratories and can support the integration of

multiple NGS instruments, it can help laboratories

track samples and optimize experimental procedures

and business workﬂows. We also give solution of in-

tegrating LIMS with the existing enterprise business

information systems. Finally, we describe the most

326

Yang J., Zhu S., Xu G. and Han C.

Laboratory Information Management System for NGS Genomics Labs.

DOI: 10.5220/0006149503260333

In Proceedings of the 10th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2017), pages 326-333

ISBN: 978-989-758-213-4

likely causes for a failed LIMS and share our experi-

ence for the implementation of a successful LIMS.

The rest of the paper is structured as follows: Sec-

tion 3 introduces the design of LIMS, while Section 4

describes the implementation of LIMS, and Section 5

presents the integration of LIMS with existing infor-

mation systems. Section 6 discusses the implemen-

tation of a successful LIMS. Section 7 concludes the

paper and outlines future work.

2 RELATED WORKS

Some commercial and open source LIMSs (Bath

et al., 2011; PerkinElmer, 2016; Grimes and Ji, 2014;

Progeny, 2016; Illumina, 2016) are available but typ-

ically require extensive modiﬁcation and extension to

address the speciﬁc needs of NGS genomics labs. We

therefore developed a web-based LIMS, which is ro-

bust and ﬂexible for managing the samples and the

NGS processes.

3 SYSTEM DESIGN

As described in Figure 1, the functionalities of our

LIMS can be grouped into the following categories:

• Enrollment of Sample Information. Enrollment

of sample information is not a simple form ﬁll-

ing process but constituted by two or even three

steps, and supports multiple business models. For

instance, salesman who stationed at hospitals will

input the basic sample information (e.g., sam-

ple code, sample type, photo of sample sheet)

into system by mobile applications embedded in

Wechat. Then, when the sample along with the

sample sheet are transported to company, typists

in company will ﬁnish the full sample information

(in sample sheet) enrollment. Finally, the correct-

ness of sample information enrolled in LIMS will

be checked by another team in company.

• Sample Logistics Information Tracking. Samples

come from different cities distributed in the coun-

try, and will be transported to our head quarter to

be tested. Each sample logistics information will

be tracked by system. When samples are packed

and sent to our company, package logistics code

will be scanned into system. Then, the system

can get the newest logistics information from ex-

press companies through their open APIs. We

can check the package logistics information status

during the whole transportation process. If some-

thing unusual (e.g., delayed, destroyed) happened

in logistics process, the recipients and senders can

response quickly to take steps to minimize the

losses.

• Sample Assessment and Processing. Depends on

sample attribute, sample sheet, sample number,

test type or more factors, the samples arrived at

company can be divided to different processing

directions including rejection, resampling, stor-

age, or ﬂow to the experiment steps. Sample re-

cipients, experiment operators and genomic ana-

lysts can choose the processing directions and op-

erate in the system.

• Experimental Management. Experimental man-

agement is the most important and complicated

part of LIMS. Based on sample type, test type,

previous processing result and operator’s subjec-

tive judgment, the samples will experience multi-

ple experimental steps such as plasma isolation,

nucleic acid extraction, molecular library con-

struction, molecular library quality control, and

etc. Our system is also ﬂexible to support the con-

ﬁguration of different experimental processes.

• Sequencing of Samples. After the experimental

steps, the samples are ready to be sequenced by

NGS instrument (e.g., Illumina HiSeq X Ten). To

minimize the cost of using sequencing reagents,

generally, one NGS instrument operation will be

required to sequencing as many samples as pos-

sible, therefore, samples belonging to different

test business line will be combined together in the

pooling process. After sequencing, the raw .bcl

data will be separated into deferent genomic anal-

ysis business processes. Abnormal result data will

be auto labeled by particular rules conﬁgured.

• Genomic Analysis. Before genomic analysis,

LIMS will prepare the raw .bcl data for the ﬁl-

ter and quality control software to be processed.

The generated FASTQ data will be further pre-

pared for the genomic analysis pipelines operated

in the cluster computing environment. After ge-

nomic analysis, LIMS will retrieve the analysis

result set to system so as to wait for the genetic

interpretation.

• Genetic Interpretation. Sequencing and analysis

results are presented in a neat and orderly man-

ner in LIMS for the genetic interpretation scien-

tists. Additionally, LIMS can provide relevant

knowledge base for genetic interpretation such

as gene-disease associations from several public

data sources (e.g. AutDB (Mindspec, 2016), Dis-

GeNET (Pinero et al., 2015), OMIM (Hamosh

et al., 2015)) and the literatures.

Laboratory Information Management System for NGS Genomics Labs

327

Enrollment of

Sample

Information

Sample

Logistics

Tracking

Sample

Assessment and

Processing

Experimental

Management

Sequencing of

Samples

Genomic

Analysis

Genetic

Interpretation

Master Data

Management

Instrument

Management

Project

Management

Query and

Statistics

System Management

Report

Management

Quality Control

Figure 1: The Application Architecture of LIMS.

• Report Management. The genetic interpretation

result will be classiﬁed by the report management

module and transferred into different report gen-

eration pipelines of the reporting system.

• Quality Control Management. Quality control

exists in the whole NGS work ﬂows, including

experiment steps quality control, reagents qual-

ity control, sequencing data quality control, ge-

nomic analysis quality control, genetic interpre-

tation quality control, and etc. Different quality

control processes can also be conﬁgured in sys-

tem.

• Master Data Management. Master data manage-

ment module includes tens of thousands of fun-

damental data for the operation of LIMS, such as

the details data for experiment templates, barcode

rules, agent information and so on. This module

is usually maintained by department manager or

system administrator, because master data include

very important and sensitive data of the company.

• Instrument Management. Instrument manage-

ment module not only records instrument prop-

erty information like barcode, type, number, loca-

tion, status, application methods and so on, but

also associates instruments to a certain experi-

ment step. An experiment step including its as-

sociated instruments could be exported as a detail

guide which is very helpful for the operators.

• Project Management. Project management mod-

ule provides the functions for project managers

a convenient way to manage and monitor the

projects, including project deﬁning, project status

monitoring, process intervening, and so on.

• Query and Statistics. Query and statistics mod-

ule provides powerful functions of advanced data

query and visualized multidimensional data statis-

tic and analysis. Instead of querying through mil-

lions of data items to ﬁnd meaningful results in

experimental workﬂows, researchers can quickly

identify the information of their interests through

this module.

• System Management. System management mod-

ule provides the features of user management, role

management, privilege management, department

management, business management, organization

management. A user’s role and his privileges

could conﬁgured ﬂexibly.

4 SYSTEM IMPLEMENTATION

Our LIMS is written in Java using the Spring (Piv-

otal, 2016), SpringMVC, Mybatis (Goodin et al.,

2016) web application framework and implementa-

tion is platform-independent. Our web servers run

Linux/CentOS and we use the MySQL relational

database management system. The system architec-

ture of LIMS is described in Figure 2.

The web interface is designed to handle a variety

of functions in a modular format. The left column

includes the function category of business process.

The top navigation bar provides advanced manage-

ment and analysis functions.

We have more than one hundred data tables in the

system, Figure 3 shows part of our database schema

for data tables.

HTTPS (Hypertext Transfer Protocol Secure) is

supported and is implemented for the system. User’s

access to functionality is controlled via user roles

HEALTHINF 2017 - 10th International Conference on Health Informatics

328

Data

Access

(DAO)

Business

Process

(Service)

Data

Control

(Controller)

View

CRM and other enterprise information systems

Oracle,

MySQL

File

System

MyBatis/

JDBC

I/O

IOS

Android

WAP

HTTP

JSP

Spring

MVC

Business

Service

Web

Service

DAO

File

Process

Web Service

Figure 2: The System Architecture of LIMS.

which are deﬁned and managed by the system man-

agement module.

Our LIMS currently supports the following Il-

lumina NGS sequencing platforms: HiSeq X Ten,

MiSeq, HiSeq, HiSeq 4000, HiSeq2500, NextSeq

500/550AR, and could be conﬁgured for other type

of NGS instruments.

5 INTEGRATION WITH

EXISTING SYSTEMS

Implementation of a LIMS requires a good degree of

integration with the existing business information sys-

tems in enterprise. We integrate LIMS to our two ex-

isting business system platforms as demonstrated in

Figure 4 and Figure 5.

5.1 Integrating with Company Systems

Figure 4 describes the business process in our com-

pany.

Customer Relationship Management (CRM) sys-

tem serves primarily our clinical and health ser-

vices, including Personal Genome Test, NIPT (Non-

Invasive Prenatal DNA Testing), Cancer Gene Ther-

apy, and etc. CRM supports sample information

collection, hospital information management, agent

management, product management, sales manage-

ment, sample management, doctor management, ﬁ-

nancial management, customer reports management,

and etc.

Project Management (PM) system serves primar-

ily our technical services, including the research area

of genomics, transcriptomics, epigenetics, and etc.

PM includes the functions of sample information

enrollment, contract management, project schedule

management, project establishment, job order man-

agement, quality control, research achievement man-

agement, and etc.

Logistics system provides the management func-

tions of express package, express company, alert time,

receipt time, samples, logistics tracking, and etc.

All the information related to NGS experiments

and analysis will be synchronized into LIMS, there-

fore, the sample information could be enrolled

through:

• CRM web portal, or

• CRM mobile app, or

• PM web portal, or

• PM mobile app, or

• LIMS directly.

The sequencing result data will be stored in Net-

work Attached Storage (NAS) system, and then ﬁl-

tered and classiﬁed to be analyzed by hundreds of dif-

ferent genomic analysis pipelines.

The genomic analysis result data will then ﬂow to

the reporting system, which can produce hundreds of

different areas of professional reports automatically.

The reporting system will ﬁnally distribute the

user reports to CRM and PM, then the user can receive

Laboratory Information Management System for NGS Genomics Labs

329

1 / 1

SampleInfo

sampleCode

testType

sampleType

hospital

outpatientCode

bloodTime

patientName

age

weight

gestationalWeeks

phoneNumber

mailingAddress

lastMenstruation

bUltrasound

trisome

inspectionDoctor

diagnosis

inspectionSheet

medicalHistory

varchar

date

varchar

UnboxCheck

unboxCode

description

courierNumber

recipient

dateReceived

status

varchar

date

varchar

PlasmaSeparation

taskLIstCode

recipient

dateReceived

bloodCode

experimentalNumber

plasmaCode

nextFlow

volume

result

varchar

date

varchar

LibraryConstruction

taskListCode

commander

orderTime

plasmaNumber

experimentalNumber

libraryNumber

originalSampleNumber

user

varchar

date

varchar

2100Test

taskCode

commander

orderTime

flow

libraryNumber

executionNumber

storageLocation

user

batchNumber

fragmentLength

quality

ratio

originalSampleNumber

molality

varchar

date

varchar

QPCRTest

taskCode

commander

orderTime

flow

libraryNumber

executionNumber

storageLocation

user

batchNumber

fragmentLength

massConcentration

ratio

originalSampleNumber

molality

varchar

date

varchar

Polling

taskCode

commander

orderTime

sequencingType

sequencingPlatform

sequencingReadLength

serialNumber

libraryNumber

executionNumber

storageLocation

originalSampleNumber

mixingRatio

conversionFactor

convertedConcentration

dilutionRatio

dilutedConcentration

mixingVolume

varchar

date

varchar

NGS

taskCode

commander

orderTime

FCNumber

reagentName

instrumentNumber

sequencingType

poolingNumber

libraryVolume

HybVolume

dilutedConcentration

degeneratedConcentration

denaturedSampleVolume

expectedDensity

NGSoperateDate

NGSStopTime

varchar

date

varchar

date

Figure 3: Part of Database Schema for LIMS.

their reports through our mobile app or web portal.

We have also constructed a hybrid cloud comput-

ing platform SolarGenomics (SolarGenomics, 2016),

because our business can produce around 10 TB se-

quencing data everyday, both the local computing

cluster and the local distributed storage can not pro-

vide computing and storage resources sufﬁciently and

elastically.

5.2 Integrating with Hospital Systems

Figure 5 describes the business process in our collab-

orating hospitals.

We collaborate with more than 2000 hospitals. In

some of the hospitals, we set up NGS laboratory and

install our NGS instrument there so that the hospitals

will have the abilities of NGS experiments, sequenc-

HEALTHINF 2017 - 10th International Conference on Health Informatics

330

CRM

LIMS

Genomic

Analysis

Reporting

System

Logistics

NGS

Instruments

app

web

Cloud

NIPTHealth Cancer

app

web

app

Genomics

EpigeneticsTranscriptomics

Figure 4: The Business Process System Platform in Enterprise.

HIS

LIMS

Cloud

Genomic

Analysis

NGS

Instruments

Samples

data

reports

Manual

Reporting

System

Head

Quarter

data

Figure 5: The Business Process System Platform in Hospital.

ing, genomic analysis, and genetic interpretation.

The samples in our collaborating hospitals could

be enrolled to LIMS either manually or synchronized

from the hospital information system (HIS). The sam-

ples will be processed by multiple experimental steps

in the laboratory of hospital and then be sequenced by

the NGS instrument placed in the hospital.

Further, the sequencing data will be:

• analyzed by our genomic analysis pipelines in-

stalled in the local hospital computing servers, or

Laboratory Information Management System for NGS Genomics Labs

331

• uploaded to our SolarGenomics cloud computing

platform then analyzed by our genomic analysis

pipelines on cloud, or

• transfered to our head quarter then analyzed by

our bioinformatics scientists.

Finally, our reporting system will give profes-

sional reports to the clinical doctors or patients.

6 A SUCCESSFUL LIMS

IMPLEMENTATION

A large number of LIMS implementations failed to

meet the user’s initial expectations, and this can be

due to the lack of proﬁcient user requirements speci-

ﬁcations, the frequent requirements changes, and the

technology-based shortcomings.

The absence of adequate requirements is the

biggest reason why a LIMS may fail, as generally the

software developers are seldom have sufﬁcient knowl-

edge of experiments and NGS, the success of a LIMS

relies on the deep understanding of laboratory and ge-

nomic analysis business needs.

Frequent requirements changes is another reason

causing the fail of LIMS, as the laboratory staffs are

normally focused on a speciﬁc area of experiment, it

is hard for them to identify clearly how a LIMS is go-

ing to ﬁt into their laboratory situation and to propose

a general function architecture of LIMS. Addition-

ally, the lab staffs lack the IT knowledge and skills,

they will think the modiﬁcation of system functions

as a simple task, therefore, the requirements will be

modiﬁed frequently. This will postpone the deliver of

project and even fail to publish LIMS.

Technology-based shortcomings should be em-

phasized, as the information technology department

is often not the ﬁrst class citizen in gene technology

organizations.

Implementation of a LIMS project will also re-

quire a good degree of integration with the exist-

ing enterprise business information systems, data ex-

change mode and standard must be adequately ad-

dressed.

Mapping out clearly the requirements speciﬁca-

tions, controlling the requirements changes effec-

tively, addressing the problem of integrating LIMS

into existing enterprise systems, relying on a strong

IT team will stand a quite high chance of implement-

ing LIMS successfully.

7 CONCLUSIONS

To meet the needs of managing tens of thousands

of samples for genome sequencing, we developed a

web-based laboratory information management sys-

tem that is ﬂexible to be conﬁgured to adapt to next

generation sequencing technologies. A LIMS system

is critical to the accurate and effective management

of sample information, experimental data, genome se-

quencing data, and the reproducible analysis results.

Our LIMS addresses all of these needs and seamless

integrates with the existing systems in enterprise and

in collaborating hospitals.

In our LIMS, all the data are stored into distributed

authoritative repositories, samples are traceable from

the enrollment, transportation, experiment, a sequenc-

ing run, quality control, genomic analysis, genetic in-

terpretation, generation of report, to sample storage,

and all the other processing steps in between. In con-

junction with a sample identiﬁer (QR code) encod-

ing rules and advanced query and analysis capabil-

ities, LIMS can quickly identify sample and signif-

icantly reduce errors in the whole steps of business

process. Our LIMS provides a comprehensive and ef-

ﬁcient management solution for NGS genomics labs.

LIMS is very complicated and difﬁcult to imple-

ment, especially in the NGS research laboratory, we

therefore share our experience of implementing a suc-

cessful LIMS.

Regarding further work, it is promising in our

schedule that we will provide a cloud-based LIMS so-

lution in our SolarGenomics genome sequencing big

data cloud platform, and open the service for more

genomics labs.

ACKNOWLEDGEMENTS

This work was partially supported by the Science

Foundation of Beijing Language and Culture Uni-

versity (supported by “the Fundamental Research

Funds for the Central Universities”) (15YJ030001,

14YBB12, 16YJ030001).

REFERENCES

Bath, T. G., Bozdag, S., Afzal, V., and Crowther, D.

(2011). Limsportal and Bonsailims: development of a

lab information management system for translational

medicine. In Source Code Biol Med.

Goodin, B., Poitras, C., and Begin, C. (2016). Mybatis. In

http://blog.mybatis.org/.

HEALTHINF 2017 - 10th International Conference on Health Informatics

332

Grimes, S. M. and Ji, H. P. (2014). Mendelims: a web-

based laboratory information management system for

clinical genome sequencing. In BMC Bioinformatics.

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

and McKusick, V. A. (2015). Online mendelian in-

heritance in man (omim), a knowledgebase of human

genes and genetic disorders. In Nucleic Acids Re-

search.

Illumina (2016). Basespace Clarity LIMS. In

http://www.illumina.com/informatics/research/

sequencing-data-analysis-management/genologics-

lims.html.

Mindspec (2016). Autdb: a Genetic Database

for Autism Spectrum Disorders. In

http://www.mindspec.org/products/autdb/.

PerkinElmer (2016). Genesifter Lab Edition. In

http://www.geospiza.com/Products/LabEdition.shtml.

Pinero, J., Queralt-Rosinach, N., Bravo, A., Deu-Pons, J.,

Bauer-Mehren, A., Baron, M., Sanz, F., and Furlong,

L. I. (2015). Disgenet: a discovery platform for the

dynamical exploration of human diseases and their

genes. In Database (Oxford).

Pivotal (2016). Spring framework. In

http://projects.spring.io/spring-framework/.

Progeny (2016). Progeny LIMS. In

http://www.progenygenetics.com/lims/.

SolarGenomics (2016). Solargenomics genome sequencing

big data platform. In www.solargenomics.com.

Laboratory Information Management System for NGS Genomics Labs

333