MICROBIOME

Analyzing -omics requires a wide range of expertise, from bioinformatics to statistical experimental design and data analysis. Having experts in multi-omics and clinical data statistical analysis on your team is critical to a project’s success. With BioRankings® you get collaborative, specialized biostatistical consultants with a focus on -omics analytics and methods.

Moving a product through the FDA requires a high level of statistical rigor to prove your hypothesis. The FDA will not accept ad hoc analyses as evidence of clinical utility. A statistical analysis plan designed before the first sample is collected that specifies statistical tests to be performed is essential to convince the FDA of trial success. At BioRankings, we do -omics based study design and formal hypothesis testing with FDA approval in mind.

Since the beginning of the Human Microbiome Project, BioRankings® statisticians and software engineers have been at the forefront for developing statistical methods, analyzing data, and developing software. This includes the first paper formally exploring sample size and power calculations for microbiome studies.

> STATISTICAL ANALYSIS PLANS AND STUDY DESIGN

A statistical analysis plan helps your team define the goals of your study. A good statistical analysis plan has clear, testable hypotheses with defined statistical tests and sample size calculations. With a strong statistical analysis plan at the start of your study, you can be sure you are sufficiently powered to discover the true signal in the data while avoiding wasting time and money on underpowered or overpowered studies. We help clients design and optimize clinical trials based on budget, required sample sizes, and goals.

 

> DATA ANALYSIS

When it is time to move forward into statistical analysis, it helps to have a partner that understands the nuances of -omics data analytics. BioRankings uses multidisciplinary approaches, drawing from a wide range of fields including statistics, machine learning, graph theory, mathematics, and computer science. In analyzing data, our toolbox includes biostatistical methods, probability theory, decision theory, and distributional assumptions.

 

> STATISTICAL METHODS DEVELOPMENT

When no methods exist for your data analysis it takes an expert to define the problem and find a solution, while avoiding false results. BioRankings staff has extensive expertise and excellent track record in inventing, testing, and implementing novel statistical methods. By applying rigorous statistical testing of algorithms before they are used with your data, you are guaranteed only validated analytical tools will be used in analysis. As an open-source company, BioRankings also provides all custom R code used.

 

> CLINICAL TRIAL PIPELINE STRATEGY

Making decisions on which products or treatments to devote time and resources to can be a challenge, especially without a clear picture of the possible limitations of past and current clinical trials. BioRankings uses your current trial data and the publicly available data from comparable studies to design and run Monte Carlo simulations to help with decision making.

Publications

 

Hypothesis testing and power calculations for taxonomic-based human microbiome data

https://www.ncbi.nlm.nih.gov/pubmed/23284876

 

Biogeography of the ecosystems of the healthy human body

https://www.ncbi.nlm.nih.gov/pubmed/23316946

 

Patterned progression of bacterial populations in the premature infant gut

https://www.ncbi.nlm.nih.gov/pubmed/25114261

 

Gut bacteria dysbiosis and necrotising enterocolitis in very low birthweight infants: a prospective case-control study

https://www.ncbi.nlm.nih.gov/pubmed/26969089

 

Exploration of bacterial community classes in major human habitats

https://www.ncbi.nlm.nih.gov/pubmed/24887286

 

Statistical object data analysis of taxonomic trees from human microbiome data

https://www.ncbi.nlm.nih.gov/pubmed/23152838

 

Combining classification trees using MLE

https://www.ncbi.nlm.nih.gov/pubmed/10204200

 

Tree-based recursive partitioning methods for subdividing sibpairs into relatively more homogeneous subgroups

https://www.ncbi.nlm.nih.gov/pubmed/11255239

 

 

(p) 314-633-1821

(a) 20 South Sarah Street

Saint Louis, Missouri 63108

© BioRankings, LLC  I  All Rights Reserved

follow us:

LEARN HOW OUR EXPERTISE CAN CHANGE YOUR BUSINESS.

I WANT MORE INFO