Computationally Driven Discovery in Coagulation

Abstract

Bleeding frequency and severity within clinical categories of hemophilia A are highly variable and the origin of this variation is unknown. Solving this mystery in coagulation requires the generation and analysis of large data sets comprised of experimental outputs or patient samples, both of which are subject to limited availability. In this review, we describe how a computationally driven approach bypasses such limitations by generating large synthetic patient data sets. These data sets were created with a mechanistic mathematical model, by varying the model inputs, clotting factor, and inhibitor concentrations, within normal physiological ranges. Specific mathematical metrics were chosen from the model output, used as a surrogate measure for bleeding severity, and statistically analyzed for further exploration and hypothesis generation. We highlight results from our recent study that employed this computationally driven approach to identify FV (factor V) as a key modifier of thrombin generation in mild to moderate hemophilia A, which was confirmed with complementary experimental assays. The mathematical model was used further to propose a potential mechanism for these observations whereby thrombin generation is rescued in FVIII-deficient plasma due to reduced substrate competition between FV and FVIII for FXa (activated factor X).

Keywords: algorithms; hemophilia A; machine learning; plasma; thrombin.

Figure 1:. Computationally Driven Approach for Discovery in Coagulation.

A mechanistic mathematical model is used to create large synthetic patient samples which are then analyzed to reveal modifiers, propose mechanisms, and generate hypotheses for experimental validation

Figure 2:. A mathematical model of coagulation under flow identifies factor V as a modifier of thrombin generation in hemophilia A.

FV is a modifier of thrombin generation in a mathematical model of flow-mediated coagulation. (A) Thrombin concentration time series generated by uniformly and independently varying plasma protein levels ± 50% from normal (110 000 total simulations); mean (solid black line), boundaries that encompass 50% (pink), and 90% of the data (orange), and the maximum/minimum of the computed solutions (gray-dashed); blue line drawn at 1nM. VAT = variance analysis time, CS = condition for samples. (B) First (blue) and total (orange) order Sobol indices are plotted as mean ± standard deviation computed with 5000 bootstrap samples of the original 110 000 simulations. (C) Plasma protein levels distributions shown as box-and-whisker plots (mean in red, whiskers drawn at three times the interquartile range), conditioned on achieving more than 1nM of total thrombin by 40 minutes. (D) Calibrated automated thrombography. FVIII deficient (<10%) plasma treated with vehicle control, 50 μg/mL exogenous prothrombin, 100 μg/mL anti-FV, and exogenous prothrombin and anti-FV. Assay conducted with 5 pM TF and phospholipids. (E) Flow assays with whole blood from FVIII deficient individuals. (A) Representative images of DiOC6 labelled platelets and leukocytes and Alexa Fluor 555 labelled fibrin(ogen) on collagen-TF surfaces at 100 per second after 25 minutes for vehicle control, 50 μg/ mL exogeneous prothrombin, 100 μg/mL anti-FV, and exogenous prothrombin and anti-FV. Scale bar = 50 μm.