Ebola virus (EBOV) infection causes a high death toll, killing a high proportion of EBOV infected patients within 7 days. Comprehensive data on EBOV infection are very fragmented, hampering efforts in developing therapeutics and vaccines against EBOV. Under this circumstance, mathematical models become valuable resources to explore potential controlling strategies. In this paper, we employed experimental data of EBOV-infected nonhuman primates (NHPs) to construct a mathematical framework for determining windows of opportunity for treatment and vaccination. Considering a prophylactic vaccine based on recombinant vesicular stomatitis virus expressing the EBOV glycoprotein (VSV-EBOV), we found that the time window can be subject-specific, but vaccination could be protective if a subject is vaccinated during a period from one week to four months before infection. For the case of a therapeutic vaccine based on monoclonal antibodies (mAbs), a single dose might resolve the invasive EBOV replication even it was administrated as late as four days after infection. Our mathematical models can be used as building blocks for developing therapeutic and vaccine modalities as well as for evaluating public health intervention strategies in outbreaks. Future laboratory experiments will help to validate and refine the estimates of the windows of opportunity proposed here.

Uncovering the hidden pathways of how antibodies induced by one influenza strain is effective against another, cross-reaction, is the central dogma for the design of universal flu vaccines. Here, we conceive a stochastic model that successfully represents the antibody crossreactive data from mice infected with H3N2 influenza strains and further validation with cross-reaction data of H1N1 strains. After modifying several aspects and parameters in the model, our computational simulations highlight that changes in time of infection and the B-cells population are relevant, however, the affinity threshold of B-cells between consecutive infections is a necessary condition for the successful Abs cross-reaction. Our results suggest a reformulation in 3-D of the current antibody influenza landscape. 1 2 3 4 5 6influenza | cross-reaction | antibody affinity | stochastic modeling | computational modeling Results 40We in-silico replicate the experimental protocol by Nachbagauer et al. (11) for cross-reactive Abs induction in mouse models. 41The experimental protocol consist of sequential infection with two divergent H1N1 or H3N2 strains of influenza virus. For H1N1 42 infections, mice were infected with an adapted human seasonal strain A/New Caledonia/20/1999 (NC99), followed by the 2009 43 human pandemic H1N1 strain A/Netherlands/602/2009 (NL09), six weeks later. NL09 is an isolate antigenically identical to 44 the prototype pandemic H1N1 strain A/California/04/09 (Cal09). The H3N2 infection was first made with a human seasonal 45 strain A/Philippines/2/1982 (Phil82), followed by another human seasonal strain A/Victoria/361/2011 (Vic11), six weeks 46 later. The viral strains in the experiments were chosen to reflect a consecutive exposure history consistent with strains that 47 Recapitulating antibodies cross-reactcome. The first H3N2 infection in Fig. 1-A, using the Phil82 strain, promotes the 56 generation of Abs of other H3N2 strains, especially of the Vict11, the A/Hong Kong/1/68 HK68 (HK68) and the A/harbor 57 seal/Massachusetts/1/11 (Mass11) strains. The response of these strains reaches the middle values of the heat-map scale of 58 Abs counts. A more attenuated response can be seen in the A/Indiana/10/11 (Indi11) and the A/canine/Texas/1/04 (Tex04) 59 strains, these strains are within the first three scales of Abs counts. On the other hand, after the second infection with the 60 Vic11 strain, practically all of the H3N2 strains reach high levels of Abs counts, except for the Tex04 strain, as shown in Fig. 61 1-B. The dynamics of Abs of a simulation case of both H3N2 infections are shown in Fig. 1-C, each strain is associated with a 62 2 | | 2Antigenically closed and distanced strains tests. We also tested the antigenic difference effect in the cross-reactive outcome by 126 selecting other H3N2 or H1N1 strains that are genetically closed or distanced each other in the genetic maps in Fig. 5. 127 Antigenically close clusters are called the internal clusters and the distanced clusters are called the external clusters...

In recent years, mathematical modeling approaches have played a central role to understand and to quantify mechanisms in different viral infectious diseases. In this approach, biologicalbased hypotheses are expressed via mathematical relations and then tested based on empirical data. The simulation results can be used to either identify underlying mechanisms, provide predictions on infection outcomes, or evaluate the efficacy of a treatment.Conducting parameter estimation for mathematical models is not an easy task. Here we detail an approach to conduct parameter estimation and to evaluate the results using the free software R. The method is applicable to influenza virus dynamics at different complexity levels, widening experimentalists capabilities in understanding their data. The parameter estimation approach presented here can be also applied to other viral infections or biological applications.Key Wo rds: viral infectio n, mathematic a l modelin g, parameter estimatio n, influen za not peer-reviewed)

No abstract

The tracking of pathogen burden and host responses with minimal-invasive methods during respiratory infections is central for monitoring disease development and guiding treatment decisions. Utilizing a standardized murine model of respiratory Influenza A virus (IAV) infection, we developed and tested different supervised machine learning models to predict viral burden and immune response markers, i.e. cytokines and leukocytes in the lung, from hematological data. We performed independently in vivo infection experiments to acquire extensive data for training and testing purposes of the models. We show here that lung viral load, neutrophil counts, cytokines like IFN-γ and IL-6, and other lung infection markers can be predicted from hematological data. Furthermore, feature analysis of the models shows that blood granulocytes and platelets play a crucial role in prediction and are highly involved in the immune response against IAV. The proposed in silico tools pave the path towards improved tracking and monitoring of influenza infections and possibly other respiratory infections based on minimal-invasively obtained hematological parameters.