Biostatistics, Epidemiology and Research Design (BERD) center 

The Biostatistics faculty and staff advise and collaborate with ACTRI researchers on biostatistics, bioinformatics, and epidemiology. In addition to common statistical tools, the faculty and staff have expertise in areas including reproducible research reporting, R Bioconductor, and high-performance computing.

The BERD Unit is supported by five full- and part-time statisticians. We have experience in a broad range of research settings, with areas of interest that include clinical trials, longitudinal data, survival analysis, observational studies, and high-dimensional data.

The BERD Unit provides statistical support at all research phases. We encourage investigators to collaborate with biostatisticians beginning in the early stages. Areas of support include:

Study Design and Proposal Writing

• Specifying and refining the research hypotheses.
• Choosing the specific aims, primary and secondary outcomes, and appropriate statistical methods.
• Reviewing the literature regarding relevant statistical methods.
• Creating randomization assignments.
• Performing sample size and power calculations.
• Writing statistical analysis plans for grants and IRB submissions.
• Responding to reviewer comments.
• Consultation on appropriate data capture, such as definition of data elements, format and structure of variables, and treatment of missing data.

Statistical Analysis

• Reviewing data sets for quality and completeness.
• Statistical software support, particularly for R and SAS.
• Fitting of statistical models; scientific computing.
• Interpreting, visualizing, and summarizing results.

Publishing and Communication

• Generating internal statistical reports.
• Editing and preparing abstracts and manuscripts for publication.
• Producing journal-quality tables and graphics.
• Responding to journal referees.

All requests must be submitted using the above "Request BERD Services" link. ACTRI membership is a prerequisite, and for grant preparation, assistance is available to  members without recharge. Up to four hours of assistance is available to ACTRI members for grant preparation without recharge. After 4 hours, you will be charged the regular hourly rate. All investigators receiving BERD assistance during grant preparation are required to include BERD costs in the budgets for their grant proposals (a practice that usually strengthens the applications).

Timelines

To ensure seamless collaboration, we encourage 6 weeks advanced notice. For requests with less than 6 weeks deadline, please contact ACTRI BERD team first. Since we usually work on multiple projects at a time, advance notice is strongly encouraged. 

• For grant submissions that do not involve preliminary data analysis, at least one month advance notice is required.
• For grant submissions involving preliminary data analysis as well as for abstract submissions, at least two months advance notice is required. It is ACTRI policy that a statistical analysis plan be formally approved by our co-investigators before analysis is carried out.

Walk-in Consultation 

Free walk-in BERD consultation is available for ACTRI members every Wednesday from 3 P.M. to 5 P.M. via zoom. Explore free walk-in BERD consultation every Wednesday from 3 P.M. to 5 P.M. via Zoom. Submit your request using the following link (https://forms.monday.com/forms/0aadcd6c446ef52837b107dd3efc0c6d?r=use1) during walk-in hours to join. Once we receive the request, we will add your name to our list and send a zoom invite to you when you are on the top of our list.

Co-authorship

Considerations of co-authorship should be based on intellectual and research contributions, which include design, analysis, and interpretation, regardless of compensation. For more on authorship guidelines, see International Committee of Medical Journal Editors uniform requirements for manuscripts.

FAQ

Who can receive support?
ACTRI membership is required in order to receive support. Please visit the membership page and apply for membership before requesting support. Investigators whose membership is pending may request support.

What happens after I submit my request for assistance? How long will it take to receive a response?
Your request will be reviewed, and a first response will be made within two work days after it has reached our BERD Unit. In the (likely) event that we have a project queue, the length of the queue typically depends on recent and current demands; the queuing time can be anywhere from a few days to several weeks. You will be informed of the specific queuing situation (if any) in our first response. When your project reaches the front of the queue, you will be contacted to arrange a face-to-face meeting. We recommend making a request as early as possible to ensure your deadlines are met (see the Timelines section above).

What will happen at the initial meeting?
The first meeting typically involves a supervising statistician, as well as one or more staff statisticians assigned to the project, and lasts no more than one hour. During the meeting, the background and research aims are discussed, and the goals of the project are outlined. Specific work items to be produced by the statistician are established (e.g., a power calculation or statistical analysis plan), including a timeline for delivery.

How will the project proceed?
After the initial meeting, the statistician will work to execute the planned objectives. Most projects will include a report detailing the results. For small projects, an analysis plan or analysis report may be sent to the investigator after the meeting; larger projects may involve multiple meetings or other communication as the investigator and statistician work together.

My deadline is this Friday, what should I do?
We are often engaged in multiple projects and generally are unable to accommodate last-minute requests. As a last resort, you may consider the weekly walk-in consultations; see above for details.

Resources

• Clinical and translational research (including educational materials)
• Sample size calculation
• Organizing data: Data Transmission Procedures or Dataset Documentation
• Importance of reproducible research

Lin Liu, PhD
ACTRI BERD Director

Dr. Liu has served as the Director of the Biostatistics, Epidemiology and Research Design (BERD) core at ACTRI since 2021. She is an Associate Professor of Biostatistics in the Division of Biostatistics and Bioinformatics at the UCSD Herbert Wertheim School of Public Health and Human Longevity Science. She has extensive experience in designing and analyzing complex observational studies and clinical trials. She has expertise in designing and validating large-scale studies using electronic health records (EHR), and developing prediction model for cancer risk classification. As an active collaborative biostatistician, she has published peer-reviewed articles in the area of HIV, cancer, gastroenterology, stroke, hypertension, infectious disease epidemiology, sleep apnea, chronic pain, psychiatric disorders including PTSD, telemedicine, and others. She is a key faculty statistician for the UCSD Moores Cancer Center with a focus on colorectal and prostate cancer research. Dr. Liu has experience training junior biostatisticians and early career physician-investigators as part of her role as the lead statistician for VA San Diego Health Services Research & Development (HSR&D). She is also the director of the Master of Biostatistics Program, mentoring undergraduates, graduate students, and postdoctoral fellows. Dr. Liu supervises post-award data analysis service and provides support to BERD liaisons in pre-award grant submissions.

Xin Tu, PhD
Professor of Biostatistics

Dr. Tu is Professor of Biostatistics and a liaison for the BERD core at ACTRI. He has been part of the BERD for five years focusing on biostatistical support primarily for pre-award grant submissions. With over 30 years of experience and over 290 publications in peer-reviewed journals, he has been making significant and important contributions to the fields of statistical methodological research and its applications to a wide range of biomedical and psychosocial research.

Xinlian Zhang
Assistant Professor of Statistics

Dr. Zhang is an Assistant Professor in the Division of Biostatistics and Bioinformatics at UC San Diego and has been a liaison for the BERD core at ACTRI since 2019. Dr. Zhang specializes in statistical approaches for big and complex data such as large-scale optimization problems, flexible and fast semi- and non- parametric methods for complex and correlated data, Bayesian spline smoothing, and longitudinal data analysis problems in statistics. Dr. Zhang also has extensive experience in analyzing and mining large-scale genomics and epigenomics data, with a specialty on RNA-seq, and ChIP-seq, metagenomic seq, and scRNA-seq.

Euyhyun Lee, MS
Senior Statistician

Euyhyun graduated from the San Diego State University with a Master of Science in Statistics in concentration of Biostatistics. He also has a Bachelor of Science in Biochemistry from the University of California, San Diego. Euyhyun joined ACTRI Biostatistics team in April of 2017 and has been collaborating with biomedical researchers on data analysis.

Khang Tong
Research Data Statistician

Khang is a graduate from UC San Diego with a Master of Science in Biostatistics where he also attained his Bachelor of Science in Applied Mathematics. Khang joined the ACTRI BERD core in 2023 where he has been collaborating with biomedical researchers on data analysis and with consultation.