How We Are Unique

At Applied BioMath, we specialize in mechanistic modeling and have extensive knowledge in the value and biological relevance it brings to mathematical modeling in drug R&D.    Our experienced team in combination with our proprietary technology enable a rigorous, high-throughput, fit-for-purpose model development process which far surpasses the capabilities of any other industry offering.

Our Modeling Approach

Our modeling approach identifies the ideal combination of mechanistic, semi-mechanistic, and statistical model components for your project to most accurately answer your scientific questions.  All of our modeling is completed in-house.  Our projects are driven by four criteria which determine the mathematical strategy for your project:

  • The scientific questions you are trying to answer
  • What type of data is available (supplied by you, from literature, or from our in-house repository)
  • Known disease and drug biology and/or mechanism of action
  • Timelines for modeling to support decision-making and project progression


We help groups answer many different types of questions depending on the specific project and the project stage.    Following is a sample of the types of questions we help answer often:

Questions typically asked early in R&D:

  • Do we target the ligand or the receptor?
  • Which step of the signaling cascade do we target?
  • What affinity do we need (e.g. 1 pM vs. 1 nM)?
  • Selectivity (e.g. 10x, 100x, 1000x) for desired target
  • How do I optimize my experimental design in the context of study limitations?
  • What are optimal drug properties?
  • Which of our preclinical candidates should we move forward?

Questions typically asked further along in R&D:

  • What is our safe starting dose?
  • What do we project an efficacious dose to be? How do we optimize the therapeutic window?
  • What patients? Indications?
  • Why didn’t my trial work?
  • Will someone come along and take this target space?
  • How do we differentiate our candidate from competitors on the market or at a later stage of clinical development?


Applied BioMath incorporates data and scientific understanding from many different types of data-sets and experimental formats, We will work with you to understand what data is available, either from your proprietary experimental data, publicly available data from literature, or from our in-house repository.   There is no required minimum amount of data for any one project. Our ability to incorporate diverse data sets combined with our biological knowledge and experience enables us to overcome gaps in particular studies.

Our collaborators may provide the following data (if available) but is not limited to:

  • Scientific input and assumptions with regards to mechanism of disease biology and drug mechanism of action
  • Client clinical data, in vitro functional and in vivo efficacy data
  • References and publications describing relevant biological mechanisms and data (including preclinical), where available, for model diagram development, model development and benchmarking

Examples might include:

  • Binding affinity data (e.g. Kd)
  • Cell binding data
  • Comparative target expression levels on various cell types
  • Quantitative receptor level expression
  • PK data
  • PD data
  • Toxicity data
  • Clinical endpoints

An important first discussion we will have is regarding what is known about your drug, target and disease biology as relevant to the project questions. Some example biological questions we might discuss are:

  • What is the drug mechanism of action (e.g. agonist, antagonist)?
  • What is the target modality (e.g. small molecule, protein therapeutic, nucleic-acid based therapeutic)?
  • What cells does it bind to?   
  • What tissues is it targeting?
  • Indication(s)?
  • Is it a large molecular or small molecule?
  • Is this a novel drug approach?


It is our goal to make deliverables and timelines as succinct as possible so you quickly see tangible results.   Prior to each collaboration, we confirm that the time we need to complete the agreed upon deliverables is suitable to meet your project deadlines.   Timelines will vary depending on the type of modeling and analysis involved in the project, but below are some example average timelines for different types of projects:

  • Traditional PK/PD modeling: typically 1 month engagements
  • Systems Models/ Mechanistic PK/PD modeling: typically 3-6 month engagements
  • QSP models/Disease models/Platform models: typically 6-12 month engagements
  • Traditional Bioinformatics (no mechanism): typically 1- 6 months
  • Large data set analysis combined with prior knowledge networks: typically 3-12 months


Industry Experience

Applied BioMath brings over 100 years of cumulative industry experience to your project.   We understand the intricacies of your project because we’ve been there.  We understand your challenges, both scientifically and within your company, and will work with you to best help you successfully move your project forward.

Prior Industry Roles

Drug Discovery Leader for Early Safety  BioMarker and Pathway Area Head, Virology  Director, Biomarkers, Infectious Disease  Clinical Pharmacokinetics  Computational Biology  Clinical Pharmacology

Prior Companies

Boehringer Ingelheim  Roche  Pfizer  Merck  Genentech  Takeda  Merrimack  Mathworks  Silver Creek  Proteostasis 

John M. Burke, PhD

John M. Burke, PhD
Co-Founder, President and CEO

Josh Apgar

Joshua Apgar, PhD
Co-Founder and CSO

Fei Hua, PhD

Fei Hua, PhD
Vice President of Modeling and Simulation Services

Proprietary Technology

Applied BioMath is constantly working to improve our technology to enable us to deliver cutting edge modeling on a time scale that keeps pace with pharmaceutical and biotechnology R&D.  Our internal modeling processes and software enable the use of advanced algorithms and scalable high performance computing without having to take shortcuts on the accuracy of the biology models.  Our technology is a continuation of work started at MIT through projects such as Kronecker Bio.  Our technology group continues to extend and develop these ideas to push the envelope of what is possible. Our goal is to never have computation limit the fidelity with which we represent the biological system or the accuracy of the scientific results we deliver.

Kronecker Bio code

jim schaff

Jim Schaff
Director of Software

Andrew M

Andrew Matteson, PhD
Senior Principal Engineer


David Hagen, PhD
Senior Principal Engineer