Establishing an infrastructure to optimize the integration of genomics into research: Results from a precision health needs assessment

Abstract

Researchers across the translational research continuum have emphasized the importance of integrating genomics into their research program. To date capacity and resources for genomics research have been limited; however, a recent population-wide genomic screening initiative launched at the Medical University of South Carolina in partnership with Helix has rapidly advanced the need to develop appropriate infrastructure for genomics research at our institution. We conducted a survey with researchers from across our institution (n = 36) to assess current knowledge about genomics health, barriers, and facilitators to uptake, and next steps to support translational research using genomics. We also completed 30-minute qualitative interviews with providers and researchers from diverse specialties (n = 8). Quantitative data were analyzed using descriptive analyses. A rapid assessment process was used to develop a preliminary understanding of each interviewee’s perspective. These interviews were transcribed and coded to extract themes. The codes included types of research, alignment with precision health, opportunities to incorporate precision health, examples of researchers in the field, barriers, and facilitators to uptake, educational activity suggestions, questions to be answered, and other observations. Themes from the surveys and interviews inform implementation strategies that are applicable not only to our institution, but also to other organizations interested in making genomic data available to researchers to support genomics-informed translational research.

We highlight the importance of integrating genomics into research, driven by a population-wide genomic screening initiative that has accelerated the need for infrastructure development in genomics research. Surveys and interviews with researchers assessed their knowledge, barriers, facilitators, and next steps for genomics research, informing implementation strategies applicable to the institution and other researchers supporting genomics-informed translational research.

Implications.

Practice: This research describes barriers, facilitators, and strategies to advancing genomics research and informatics.

Policy: Policymakers working to advance quality and accessibility in genomics care and cancer prevention should advocate for the frameworks and strategies developed through this research.

Research: These frameworks are generalizable to researchers investigating genomics-informed translational research, and future research should be aimed at identifying procedures for optimizing delivery of care and care guidelines.

Introduction

The rapid pace of genomic discovery has created an unprecedented opportunity to integrate genomics across the translational research continuum. Genomic research ranges from basic science (T0 research) that involves gene discovery to individualized health applications that could be used in clinical and public health settings through population-focused application (T4 research), including the adoption of evidence-based recommendations and guidelines to support health outcomes. While genomic discovery has predominately occurred in basic science research, there are growing opportunities to incorporate genomic findings for clinical and public health benefit [1, 2].

Population-level genomic research programs have expanded dramatically, furthering the ability for researchers to access genomic data. While the National Institutes of Health All of Us Research Program continues to enroll individuals to provide biospecimens to their national cohort, population-wide genomic screening studies are becoming more common among health systems and research institutions [3]. Williams reported on at least a dozen population-wide genomic screening programs in the USA, which are designed to offer both clinically relevant information (e.g. results for CDC Tier 1 conditions: Hereditary Breast and Ovarian Cancer, Lynch Syndrome, and Familial Hypercholesterolemia) and be used to develop a research database for researchers within a specific health system or institution [4]. These initiatives are aligned with the National Human Genome Research Institute’s 2020 vision of “Compelling Research Projects,” which are a result of integrating basic genomics research findings with genomic learning healthcare systems to generate new knowledge [5]. The proliferation of these databases and analytic tools have provided unprecedented opportunities for discovery but require substantial knowledge and training to be used to their fullest extent.

As genomic data becomes increasingly available to researchers across the translational research continuum, the research community must be adequately prepared to integrate genomic data to ensure the full benefits of genomic discovery are realized [6]. While there has been substantial focus on advancing genomic medicine and preparing clinicians to integrate genomics into their clinical care, there has been less emphasis on how to train and prepare researchers in various field to begin incorporating genomics to support their research programs (e.g. genome-wide association studies, phenome-wide association studies) and ensuring they have the appropriate tools, resources, and skills to conduct such studies [7–9].

In Our DNA SC is a large-scale population-wide genomic screening initiative at the Medical University of South Carolina (MUSC) has a goal of recruiting 100 000 total participants. Like other population-wide genomic screening studies, genomic data from participants is linked to participant’s electronc health record data and accessible to researchers at our institution interested in using the dataset for novel research. We conducted a needs assessment to help understand facilitators and barriers to using this dataset and to inform the strategies necessary to appropriately prepare investigators to use the data. The specific goals of this needs assessment were to (i) understand the specific needs of researchers at MUSC related to using genomic data for precision health research, and (ii) identify how best to provide effective education and training support. Lessons from this effort also grow the literature available and can be applied to other programs seeking to expand capacity for their institution to optimize research efficacy in precision health.

Methods

The evaluation of the In Our DNA SC program was approved through the MUSC IRB under protocol 0011732. A Precision Health Needs Assessment was developed for the purpose of this investigation and was composed of a survey and follow-up qualitative interviews. The survey was developed by our study team and distributed through research-focused list servs at MUSC. The survey included questions about the responding individual’s current research program (e.g. type of translational research, past or current funding related to genetics or genomics). We asked individuals to rank the importance, relevance, value, priority, frequency, and likelihood of the use of genomic data in the researcher’s specific field or line of research (5-point Likert scale). We asked individuals to indicate perceived barriers and facilitators to using genomic data (e.g. complexity of genomic data, compatibility of genomic data to research, organizational incentives, and rewards to use genomic data, institutional resources to support use of genomic data, institutional infrastructure to use genomic data, access to genomic data, access to education about genomic data, personal knowledge about genomic data, and personal capabilities to use genomic data) (5-point Likert scale). We asked individuals about their level of familiarity with various tools related to research (completing research data requests, IRB submission, SQL server management studio, self-service data access tools, large-scale data management, large-scale data analysis, use of de-identified, or aggregate data for research, use of identified data for research, participating in multisite studies, and obtaining data sets for common data models) (5-point Likert scale). We also asked about preferred training formats for future educational initiatives. At the conclusion of the survey, we asked individuals whether they would be interested in participating in a follow-up interview to further discuss precision health and genomics.

Individuals who indicated interest in participating in a follow-up interview via survey were interviewed by a member of the Office of the Vice President of Research at MUSC. 30-minute interviews were conducted virtually with the interviewer and individual participant and were recorded and transcribed for later review. Interview questions were developed based on prior research conducted by the study team [10]. This script included six conversation-guiding questions was used by the interviewer. From questions asked via the script, each participant provided a summary of their research perspective and goals, as well as their interest in integrating precision health into their work. Questions expanded upon the quantitative survey and included: alignment of current research with precision health and genomics, opportunities to incorporate precision health into their research, barriers, and facilitators to uptake of genomics in research, and educational activity suggestions to help researchers learn about genomics.

Data analysis

Quantitative survey data were analyzed using descriptive statistics (mean, standard deviation, or frequency, percent). Qualitative data from participants were audio-recorded and transcribed. A summary was created immediately following the interview to capture key points and assist with codebook development. A list of codes was developed by the study team based on the semi-structured interview guide. Two members of the study team independently coded each interview and disagreement in assignment or description of codes was resolved through discussion between investigators or through modifying code definition. Researchers used a rapid qualitative analysis approach to code the transcripts, involving two researchers cross-validating coding methods through addressing and modifying any disagreement in code assignment. Themes and supporting quotes for each code were pulled. After individual interviews were coded, interviews were collectively analyzed to assess overarching themes, questions, and ideas. We completed data triangulation using a mixed methods matrix during the analysis phase, integrating rapid qualitative analysis elements as well as descriptive statistics regarding quantified survey responses.

Results

Participants

A total of 36 individuals completed the quantitative survey (Table 1). Those who participated in the survey came from various affiliations across MUSC: College of Dental Medicine (n = 6, 16.7%), College of Graduate Studies (n = 1, 2.8%), College of Health Professions (n = 3, 8.3%), College of Medicine (n = 22, 61.1%), College of Nursing (n = 1, 2.8%), College of Pharmacy (n = 1, 2.8%), and Other (n = 2, 5.6%). Their titles varied from Assistant Professor (n = 12, 33.3%), Associate Professor (n = 7, 19.4%), Professor (n = 12, 33.3%), and to Other (n = 5, 13.9%). Respondents’ research programs occurred across the Translational Research Continuum (participants were able to choose multiple levels, if applicable): T0: Basic Science Research (n = 13, 36.1%), T1: Translation to Humans (n = 6, 16.7%), T2: Translation to Patients (n = 11, 30.6%) T3: Translation to Practice (n = 16, 44.4%), and T4: Translation to Community (n = 10, 27.8%). Fourteen (38.9%) respondents received prior funding related to genetics or genomics and n = 4 (11%) taught a course relevant to genetics or genomics.

Table 1.

Sociodemographic of survey and interview participants

	Survey		Interviews
	N	%	N	%
Affiliation
College of Dental Medicine	6	16.7	0	0
College of Graduate Studies	1	2.8	0	0
College of Health Professions	3	8.3	0	0
College of Medicine	22	61.1	6	75
College of Nursing	1	2.8	0	0
College of Pharmacy	1	2.8	1	12.5
Other	2	5.6	1	12.5
Title
Assistant Professor	12	33.3	3	37.5
Associate Professor	7	19.4	3	37.5
Professor	12	33.3	1	12.5
Other	5	13.9	1	12.5
Translational research continuum
T0: Basic Science Research	13	36.1	2	25
T1: Translation to Humans	6	16.7	2	25
T2: Translation to Patients	11	30.6	2	25
T3: Translation to Practice	16	44.4	5	62.5
T4: Translation to Community	10	27.8	4	50
Prior funding
Yes	14	38.9	3	37.5
Teach relevant course
Yes	4	11	2	25

Eight of the 36 survey completes elected to participate in the qualitative follow-up interview (Table 1). Interview participants reflected a variety of backgrounds including but not limited to Alzheimer’s research, cardiology, molecular pathology, proctology, predictive pharmacology, psychology, pulmonary critical care, translational science, and tobacco treatment. Their affiliations were n = 6, 75% from the College of Medicine, n = 1, 12.5% from the College of Pharmacy, and n = 1, 12.5% from Other. Their titles varied from Assistant Professor (n = 3, 37.5%), Associate Professor (n = 3, 37.5%), Professor (n = 1, 12.5%), and to Other (n = 1, 12.5%). Interviewees’ research programs occurred across the Translational Research Continuum (selecting multiple levels, if applicable): T0: Basic Science Research (n = 2, 25%), T1: Translation to Humans (n = 2, 25%), T2: Translation to Patients (n = 2, 25%), T3: Translation to Practice (n = 5, 62.5%), and T4: Translation to Community (n = 4, 50%). Three interviewees (37.5%) had received funding previously related to genetics or genomics and n = 2 (25%) taught a relevant course.

Value of genomics

When asked to rank the importance, relevance, value, priority, frequency, and likelihood to use genomic data in the researcher’s specific field or line of research (Likert scale 1–5, where 1 was “not at all,” 3 was “neutral,” and 5 was “very”), relevance was ranked highest (M = 3.9, SD = 1.1) (Table 2). Value of genomics within the researcher’s specific field or line of research was ranked second highest (M = 3.8, SD = 1.1), followed by importance and likelihood to use genomics data (M = 3.6, SD = 1.3). Priority of genomic use within the researcher’s specific field or line of research was ranked second to last (M = 3.4, SD = 1.3) with frequency with which you use genomic data for your specific field or line of research ranking last (M = 2.7, SD = 1.5).

Table 2.

Results from precision health needs assessment

Value of genomics	Mean	SD
Importance of genomics to specific field or line of research	3.6	1.3
Relevance of genomics to your specific field or line of research	3.9	1.1
Value of genomics to specific field or line of research	3.8	1.1
Priority of genomics to your specific field or line of research	3.4	1.3
Frequency with which you use genomic data for your specific field or line of research	2.7	1.5
Likelihood to use genomics for your specific field or line of research	3.6	1.3
Level of familiarity	Mean	SD
Completing SCTR SPARC requesta	3.8	1.2
IRB submission	3.8	1.3
SQL server management studio	1.6	1
Self-service data access tools	2.1	0.9
Large-scale data management	3.1	1.4
Large-scale data analysis	2.9	1.4
Using de-identified or aggregate data for research	3.4	1.3
Using identified data for research	3.2	1
Participating in multisite studies	3.1	1.4
Obtaining data sets for common data models	1.9	1.2
Preferred training format	N	%
Hands on workshops	26	72.2
Group session	15	41.7
One-on-One session	15	41.7
Lecture	16	44.4
Online tutorial	17	47.2
Webinar	16	44.4
Other	2	5.6

^aA web-based research management system that provides a central portal to researchers and their study teams to browse for research services and resources.

Interviews confirmed these themes, as all interviewees expressed interest in integrating precision medicine and using a centralized genomics database in their research (Table 3). Interviewees expressed that although other large genomics databases exist, the database generated through In Our DNA SC would be more representative and predictive of the local population. One researcher indicated that genomics data was the “missing piece” to their research, a sentiment echoed by others.

Table 3.

Qualitative findings

Topic	Themes	Quotes
Alignment of research with precision health	Awareness about ways to use of genetics to gain more insight about their translational research Although not using genomic data, recognize opportunities to build into existing research agenda Clear alignment and interest in specific genes associated with research topic	“I don’t really do genetic research like that is not part of what I do on a day-to-day basis, but in the literature like there’s a lot of opportunity especially in the populations that I work with.” “One of the reasons that we had been collecting samples in which to do DNA testing is because there’s a particular Alzheimer's disease, late onset Alzheimer’s disease risk, gene called APOE.”
Opportunities to incorporate precision health	Access to large-scale databases can help with specific research questions Personalization of medical treatments and dosing of medicines Genetics viewed as missing piece of data and necessary to incorporate Need to improve representation of diverse participants for research cohorts Importance of genomics and data for grants and funding	“They have extensive imaging like there’s a ton of data that can be mined. You know. But you know, we just kind of have that missing genetic piece.” “I think a lot of that could be done digitally. I think it would help patients get the right drugs at the right time. I think it would reduce costs of giving drugs at the wrong time to the wrong people, as well as the safety issues of that.”
Barriers to uptake	Multidisciplinary workforce is needed and requires diverse education and mentorship Selecting genes and topics that are familiar rather than what is relevant Relevance of findings to patient populations Different levels of comfort with genomics Data storage and security General datasets may not be relevant to research of interest Cost related to storage, expertise (statistical expertise)	“The more complicated answer is different people will benefit or not benefit each in their individual ways, so coming up with the right coursework for example, that addresses everybody is going to be complicated.” “And then you quickly realize that this is a career just like any other career where you may specialize in what you know [...] And so one of the things that we are learning is that there are various levels of comfort with genomics period. Like what does all this mean, and what does it mean? What does all this mean? Number one, and then #2, how do I extrapolate that to what it means for my research study or my line of inquiry?” “I think a lot of people when they, you know, deal with genomic level information maybe don’t have the best understanding of you know the uses of that type of information over time”
Educational activity suggestions	Succinct resources and instructions Offer personalized, 1-1 support in real-time Support with data interpretation Presentation about research within existing meetings so collaborators can find each other Link researchers with different backgrounds and expertise together	“I think I’m already sophisticated enough to know how to understand and use the information that will be assembled as long as the interface is made user friendly” “I think it’s still is data interpretation help. I think a lot of people are gonna have a kind of question in their head, but aren’t gonna be quite sure how to ask that question to the data set. More than that, I think they’re gonna need to be properly educated and coached on what the data actually means”

Interviewees all expressed a high demand for a genomic data and a large perceived benefit for integrating genomics into their work. Primary facilitators for adoption included that researchers would save energy, financial resources, and time by using the data provided through In Our DNA SC instead of collecting the data themselves. Additionally, the ability to use the network of collaborators, mentors, and consultants to fill any knowledge gaps or provide guidance on a project proved a large incentive for adoption. Researchers also saw benefit to bringing clinical and genomic data together into a centralized resource. As a result, interviewees expressed a high level of commitment to use genomic and genetic data in their research program. Additionally, interviews indicated that implementation of precision health would improve researcher self-efficacy by providing the resources to educate themselves, view examples of other research developed from the database, and easy ways to obtain assistance when necessary.

Barriers and facilitators to use

The largest perceived barriers to use of genomic data were the complexity of genomic data (n = 30, 83.3%) followed by internal infrastructure (n = 28, 80.0%) and internal resources to support use (n = 27, 77.1%) (Table 4).

Table 4.

Barriers and facilitators to using genomic data

	Barrier		Facilitator
Barriers and facilitators to use	n	%	n	%
Complexity of genomic data	30	83.3	6	16.7
Compatibility of genomic data to your research	16	47.1	18	52.9
Organizational incentives and rewards to use genomic data	19	57.6	14	42.4
Internal resources to support	27	77.1	8	22.9
Internal infrastructure	28	80.0	7	20.0
Access to genomic data	26	76.5	8	23.5
Access to education or information	24	75.0	8	25.0
Personal knowledge about genomic data	23	65.7	12	34.3
Personal capabilities to use genomic data	27	75.0	9	25.0
Missing frequency ranged from 1 to 3

Perceived facilitators that were identified included: compatibility of genomic data to respondent’s research (n = 18, 52.9%), organizational incentives, and rewards to use genomic data (n = 14, 42.4%), and personal knowledge about genomic data (n = 12, 34.3%).

We found notable consistencies regarding barriers and facilitators to using genomic data for precision health research in our qualitative interviews. A lack of the technical skills to use the database was a commonly noted barrier. Given that researchers have diverse skill sets and varied research and educational needs, a universal educational tool was not suggested. Instead, researchers requested optional, specialized educational workshops, and collaboration opportunities. However, proper usage of the dataset was a continued concern, with researchers echoing that such large datasets with simple algorithms leave substantial room for misinterpretation. In other words, statistics for these datasets can be difficult to understand and easy to inadvertently manipulate, so extrapolating meaning from these datasets is challenging. Ensuring that researchers are aware of their access to the dataset, and storing the data in a way that allows easy access and sharing of findings were also concerns. Interviewees suggested training advisors to help connect researchers with mentors and consultants to fill gaps in their research knowledge. These advisors, along with examples of previous research questions and outcomes, could help researchers develop informed research questions and plans. Respondents indicated that organized and centralized resources would promote synergy between databases for research and clinical usage and may even encourage clinicians to enroll patients in research.

There were several challenges identified to successful integration of precision health into current research practices. First, educating a busy and diverse workforce with a wide range of experience and expertise is difficult. One researcher summarized the need for effective educational materials sharing, “One of the things that we are learning is that there are various levels of comfort with genomics, period. Like what does all this mean? And then #2, how do I extrapolate that to what it means for my research study or my line of inquiry?” However, by breaking down the educational categories and offering both generalized and tailored support at the relevant times, efficiency can be optimized without compromising quality, allowing researchers to customize their education to their specific gaps in knowledge. Another common concern was researchers being deterred from or improperly using the dataset due to lack of expertise in statistical analysis and data interpretation in genomics. One researcher emphasized this issue, sharing “Interpreting and understanding statistics in that kind of a large space is nontrivial. And one of the problems is some of the search algorithms make it feel trivial.” However, by pairing the database access with mentor matching and support from statistical geneticists, the barrier to entry for using the database can be lowered and the level of experience and expertise collaborating on these projects increased. One researcher also noted that the availability of such a large dataset may make publishing new research easy but does not guarantee improvements will be implemented at the patient level. This concern is one reason for emphasizing collaboration, not only across specialties but also between clinical and nonclinical researchers.

Level of familiarity

Respondents were asked about their level of familiarity with different tools that could be utilized to access genomic data (Table 2). The tools that were identified as most familiar to researchers included: completing a South Carolina Clinical and Translational Research (SCTR) Services, Pricing, and Application for Research Centers (SPARC) Request (M = 3.8, SD = 1.2), completing an Institutional Review Board (IRB) submission (M = 3.8, SD = 1.3), and using de-identified or aggregate data for research (M = 3.4, SD = 1.3). Tools for utilizing genomic data that respondents were least familiar with included SQL server management studio (M = 1.6, SD = 1), obtaining data sets for common data models (M = 1.9, SD = 1.2), and self-service data access tools (M = 2.1, SD = 0.9).

Many interviewees had some familiarity with genomics databases. All expressed motivation to advance integration of precision medicine into their work, especially an easily accessible database representative of their local patient population.

While the applications of genomic databases were often specialty-specific, some opportunities to incorporate precision health transcended specialty. For example, researching biomarkers for relevant diseases, improving classification, and specification of diseases, conducting disease pathway analysis, RNA sequencing, pharmacogenetics, predictive pharmacology, and drug metabolism were interests found throughout specialties. For example, one researcher emphasized the opportunity of the database to allow for personalizing type, dosage, and timing of medicine based upon genetics. This specialist shared that this customization “would help patients get the right drugs at the right time. I think it would reduce costs of giving drugs at the wrong time to the wrong people, as well as the safety issues of that. And I think that would help provide better access both to healthcare and the right drugs through a more efficient pipeline.” Researchers anticipated using the database to help design further research studies, improve ethnic and racial representation in their work, and to merge clinical and nonclinical datasets and research.

Preferred training format

Respondents were asked their preference on a broad range of training formats to learn about genomic research (Table 2). Notably, hands on training (n = 26, 72.2%) was the preferred training format, followed by an online tutorial (n = 17, 47.2%), with either a lecture or a webinar (n = 16, 44.4%) as the next preferred format for learning about genomics research. Fifteen survey respondents (41.7%) indicated that they would prefer a group training session or a one-on-one session.

Researchers indicated what support was needed for them to make the most out of the genomic database. Having an easily accessible and searchable database with sharable datasets was important to interviewees. Interviewees requested access to previous research studies that used the database to serve as an example and inform further research designs and questions. Providing a system for connecting collaborators and mentors was also an incentive for adopting genomics, especially among less-experienced researchers. Interviewees agreed education on how to use the database was necessary. Opinions on best methods and content preferences were thoroughly discussed. Generally, four types of education were discussed: informing faculty of the existence of and access to the database and resources, instructions regarding database user interface and how to access information in the database, technical skills needed to extrapolate meaning from such a large and complex database, and lastly legal, ethical, and business implications of genomics research. To ensure faculty is aware of access to the database, respondents recommended promoting the use of genomic data at university events and conferences, as well as advertising access in newsletters. Researchers had differing opinions on ideal education strategies for how to use the database. The most common viewpoint was that researchers should be primed with educational videos on basic, easy database functions, then offer optional small group workshops to ask questions, and advance skillsets that can be attended in person or online. Additionally, interviewees advocated for support tools to be built into the dataset to support work in real-time. Educating researchers on the technical skills needed to extrapolate meaning from the database was one of the most consistently voiced concerns represented in the interviews. Researchers advocated for assistance with data interpretation and coaching on developing research questions and plans. One researcher explained the importance of this technical support for researchers sharing, “I think a lot of people are going to have a kind of question in their head, but aren’t going to be quite sure how to ask that question to the dataset. More than that, I think they are going to need to be properly educated and coached on what the data actually means to them.” The proposed solution was one-on-one consults for researchers with statistical geneticists, mentors, and other relevant experts to ensure proper usage of the database through the duration of their research project. Lastly, respondents indicated that training in eugenics, relevant HIPAA protocols, and associated laws such as the GINA act, the Omnibus, and the HITECH Act should be provided as a succinct online learning module. One specialist in predictive pharmacology and drug informatics also advocated for education of Principal Investigators on business skills, grant information, and patent processes, as these are highly relevant for that specialty and may increase motivation to pursue genetic and genomic research projects. These materials would be presented as optional asynchronous online modules.

Discussion

We conducted a precision health needs Assessment during early development of research infrastructure at MUSC to support precision health research using a genomic database. The goal of the assessment was to better understand the potential impact of a genomic database on research as well as approaches to improve adoption and success of the program.

Results from quantitative surveys and qualitative interviews revealed clear interest in integrating genomics into research programs. Notably, respondents sought to use the In Our DNA SC database specifically because it will provide access to genomic data representative of the population of the state of South Carolina. Other opportunities to enrich the dataset included recruitment of specific cohorts of interest to researchers at the institution (e.g. individuals with certain clinical indicators). Researchers indicated that centralizing genomic data collection through the In Our DNA SC project was a substantial strength of the program, as it reduced barriers for investigators new to genomics research by eliminating the need to collect genomic data for their specific study.

Despite these motivations, our needs assessment revealed concerns about proper use of the database. Qualitative findings suggested that the complexity of this type of dataset may make it difficult to understand and use, possibly resulting in inadvertent manipulation or drawing inappropriate conclusions. In response to this concern and in alignment with best practices for genomic data use, the In Our DNA SC program has established a Research Governance Committee. The committee consists of nine individuals with expertise in human participant research, human genomics, ELSI implications of genomic research, health disparities research, data privacy and security, and data science. The goal of the Research Governance Committee is to assess the objectives of genomic research requests to ensure that participant data will be leveraged in a responsible way to advance precision medicine research and fuel new insight into human health that benefits all. The responsibilities of the committee include the development and implementation of policies for data access, review of research proposals, and review of potential violations of the Data User Code of Conduct. Additional support is available to investigators through one-on-one consultations with the Research Governance Committee to discuss potential research questions and opportunities to integrate genomics into their research program.

The complexity of genomic data and limited self-efficacy for using the dataset were among top barriers to using the In Our DNA SC database. While respondents were able to conceptualize potential research questions related to their research programs, they did not have the appropriate skills to move from the question to a research program plan of action with the dataset. Specific recommendations to mitigate barriers included: development of templates for data management, expert consultations, providing examples of research from across the translational research continuum, and champion dyads where individuals who are experts in the area are paired with more novice researchers. Other projects have found success in developing practical toolkits and implementation guides to support the integration of genomics into clinical care [8, 11–13]. This approach could be replicated to support the integration of genomics into research across the translational research continuum.

To address the challenges identified during this needs assessment, the In Our DNA SC program is developing technical resources and educational materials including a series of six modules to engage researchers in the use of genomic data. Topics range from basic concepts and research questions to technical aspects of requesting and analyzing genomic data for both clinical and complex traits applications. These modules were delivered in May and June 2023, and will be evaluated to identify opportunities to continue enhancing education for researchers. While these modules and resources are designed to be directly responsive to the challenges raised during the needs assessment, they also highlight the importance of supporting transdisciplinary research teams. Enabling genomics research will not only involve equipping researchers with basic genomic knowledge, but it will also, importantly, support opportunities for researchers to connect and form new research partnerships to address novel research questions. One of the greatest gaps identified in our findings is a disconnect between the high perceived value of genomic data in enhancing investigator’s current work and a lack of basic knowledge about getting starting this research and performing it well. Specific recommendations included genetics consultants who could guide investigators in research planning, covering aspects such as tools, question design, and statistical considerations. The Research Governance Committee is currently providing support to investigators interested in using the In Our DNA SC data as part of their research; however, additional modalities such as formal consultations and training may be necessary to fully support investigators new to using genomic data [6, 10, 12].

This study is not without limitations. Data were from a cross-sectional sample of researchers at one institution, which limits the generalizability of our findings. Contextual factors such as our new population-wide genomic screening study, specific institutional resources, and setting were not accounted for as part of the needs assessment. Multi-institutional surveys could help improve generalizability and account for multilevel factors that impact researcher needs when developing a genomic research agenda. Additionally, the individuals who participated in the qualitative interviews may not necessarily be representative of all investigators at the institution or across other institutions, limiting the generalizability of our findings. Finally, due to our limited samples size we did not conduct subgroup analyses or comparisons across multiple groups (e.g. early career vs. later career investigators, clinical investigators, or other investigators). Future work could consider variation in subgroups.

Through this mixed methods needs assessment, we identified barriers and facilitators of integrating genomics into translational research at our institution. These findings informed institutional next steps to improve access and use of genomics in research. Our efforts can serve as a model for smaller institutions that lack an established genomics infrastructure seeking to build a precision medicine research infrastructure to leverage populations to promote diversity, equity, and inclusion for precision medicine research. Genomic information will continue to become ubiquitous in both clinical and research settings. Establishment of policies, resources, and partnerships are necessary to appropriately prepare investigators to integrate genomics across the translational research continuum.

Building upon the insights from the needs assessment, our future research plans involve implementing comprehensive resources and modules designed to educate and engage researchers in the use of genomic data. Additionally, we aim to evaluate the effectiveness of these educational resources, seeking opportunities for continuous improvement and refinement. Our commitment to fostering transdisciplinary research teams will continue, emphasizing the importance of supporting researchers in forming new partnerships to address novel research questions. In light of existing literature, our findings underscore the growing demand for tailored educational initiatives and specialized support structures for researchers aiming to incorporate genomics into their work. The barriers identified, such as the complexity of genomic data and the need for internal resources and infrastructure, align with broader discussions in implementation science regarding the challenges of integrating novel technologies into existing workflows.

Our contribution to genomics literature lies in the practical insights derived from the needs assessment, offering a blueprint for institutions seeking to expand their genomic research capacity. As population-level genomic initiatives become more prevalent, our study provides timely guidance on how to prepare and support researchers in leveraging large-scale genomic datasets for precision health research.

Funding

Allen was supported by K00CA253576.

Conflict of interest statement. None declared.

Ethical Approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed Consent

Informed consent was obtained from all individual participants included in the study.

Transparency Statements

(1) Study registration: Not formally registered. (2) Analytic plan preregistration: Not formally preregistered. (3) Data availability: De-identified data are not available in a public archive. (4) Analytic code availability: Analytic code used to conduct analysis are not available in a public archive. They may be available by emailing the corresponding author. (5) Materials available: Materials used to conduct the study are not publicly available.