Daniel Scholes Rosenbloom

Harvard University | Harvard · Program for Evolutionary Dynamics

Background Allogeneic blood or marrow transplantation (alloBMT) is a potentially life-saving treatment for individuals with HIV and haematological malignancies; challenges include identifying donors and maintaining antiretroviral therapy (ART). The objectives of our study were to investigate interventions to expand donor options and to prevent ART...

Background: Evaluations of HIV curative interventions require reliable and efficient quantification of replication-competent latent reservoirs (LR). The "classic" quantitative viral outgrowth assay (QVOA) has been regarded as "gold standard," although prohibitively resource- and labor-intensive. We compared six "next-gen" VOA employing PCR or ultr...

Quantitative viral outgrowth assays (QVOA) use limiting dilutions of CD4⁺ T cells to measure the size of the latent HIV-1 reservoir, a major obstacle to curing HIV-1. Efforts to reduce the reservoir require assays that can reliably quantify its size in blood and tissues. Although QVOA is regarded as a “gold standard” for reservoir measurement, litt...

Laboratory protocols. (PDF)

QVOA well configuration. (XLSX)

Experimental setup, showing lab and batch assignment of each aliquot. (XLSX)

Complete experimental dataset. (XLSX)

Performance of MCMC estimation using the model with positive aliquot- and batch-level variation and HC(0,2) prior, in single-lab simulation. (PDF)

Performance of MCMC estimation using the ensemble model and HC(0,4) prior, in multi-lab simulation, as in Table 2. (PDF)

Performance of MCMC estimation assuming a true random effect parameter of zero, using HC(0,1) prior, in multi-lab simulation. (PDF)

Summary of model fits. Parameter distributions of the ensemble model are reported as weighted quantiles of 8000 posterior samples, and the number of samples in the discrete ensembles are shown (see Methods: “Markov-chain Monte Carlo estimation”). Analysis is shown for eight different subsets of the data: three subsets including multiple labs, and f...

Pairwise comparisons of assay accuracy for higher IUPMs, assuming that each assay has equal claim to biological truth (σc applied to both assays). Method and format match S12 Table. (PDF)

Performance of MCMC estimation using the ensemble model and HC(0,2) prior, in multi-lab simulation, as in Table 2. (PDF)

Performance of MCMC estimation using the ensemble model and HC(0,3) prior, in multi-lab simulation, as in Table 2. (PDF)

Pairwise comparisons of assay accuracy for lower IUPMs, assuming that each assay has equal claim to biological truth (σc applied to both assays). Batch variation-free ensemble estimates of parameters were used in simulations. Each entry shows median improvement of row versus column assay (median and 95% CI sampling from ensemble parameter distribut...

Pairwise comparisons of assay accuracy for lower IUPMs, treating the JHU assay as gold standard (2×σc applied to other assays). Method and format match S12 Table. (PDF)

Pairwise comparisons of assay accuracy for higher IUPMs, treating the JHU assay as gold standard (2×σc applied to other assays). Method and format match S12 Table. (PDF)

Simulated results of latency reduction trials, maximum likelihood estimation to compare pre- and post-treatment data. (PDF)

Sensitivity analysis of of aliquot-level, batch-level, and lab-level variation. (TIF)

Accuracy of assays used in the experimental study, with JHU and SR assays modified to use the same number of input cells as U. Pitt. and UCSD. Each assay is measured against a consensus standard, appropriately scaled by βl for that assay. “All-negative” on the y-axis represents infinite error on the fold-change scale, which occurs when the maximum...

Stan model code. Complete model specifications are provided in the Stan programming language (version 2.12) as supplementary text files. Each filename has a binary code, indicating presence/absence of aliquot and batch effects (for single-lab models) or presence/absence of aliquot, batch, and lab effects (for multi-lab models). In the code, SourceI...

Scatterplot of estimates of excess variation at the aliquot, batch, and lab levels, expressed as the natural log of fold-change (parameters σa, σb, σc). Panels show robust linear regression and Wald test p-values (null model = zero slope) of 1000 samples from the ensemble posterior. (TIF)

Outcomes of anticancer therapy vary dramatically among patients due to diverse genetic and molecular backgrounds, highlighting extensive intertumoral heterogeneity. The fundamental tenet of precision oncology defines molecular characterization of tumors to guide optimal patient-tailored therapy. Towards this goal, we have established a compilation...

The odds of living a long and healthy life with HIV infection have dramatically improved with the advent of combination antiretroviral therapy. Along with the early development and clinical trials of these drugs, and new field of research emerged called viral dynamics, which uses mathematical models to interpret and predict the time‐course of viral...

Quantitative viral outgrowth assays (QVOA) use limiting dilutions of CD4+ T cells to measure the size of the latent HIV-1 reservoir, a major obstacle to curing HIV-1. Efforts to reduce the reservoir require assays that can reliably quantify its size in blood and tissues. Although QVOA is regarded as a “gold standard” for reservoir measurement, litt...

A topological approach to the study of genetic recombination, based on persistent homology, was introduced by Chan, Carlsson, and Rabad\'an in 2013. This associates a sequence of signatures called barcodes to genomic data sampled from an evolutionary history. In this paper, we develop theoretical foundations for this approach. First, we clarify the...

Viral infections are one of the major causes of death worldwide, with HIV infection alone resulting in over 1.2 million casualties per year. Antiviral drugs are now being administered for a variety of viral infections, including HIV, hepatitis B and C, and influenza. These therapies target a specific phase of the virus’s life cycle, yet their ultim...

Average equilibrium infection level for the multi-strain competition deterministic model with pharmacological drug kinetics, as a function of the average maturation time (1/m). The infection level is measured as the concentration of mature infected cells y. (a) Results of deterministic model with n = 10 maturation steps. (b) Results of deterministi...

Life cycle data and example drug treatments for common viruses. (TIFF)

Average equilibrium infection level for the multi-strain competition deterministic model with imperfect therapy adherence and on/off drug kinetics, as a function of the average maturation time (1/m). The infection level is measured as the concentration of mature infected cells y, under anti-viral therapy drug efficacy f = 0.85 and period T = 2 days...

Time course of early infection dynamics and calculated growth rate, r˜0. Infection dynamics with the fixed-delay model (Eq (3)) were numerically integrated starting from a small amount of infection introduced to the uninfected equilibrium. The level of mature infected cells is shown (blue line). To estimate r˜0 from these dynamics, we took points s...

Supplementary methods. (PDF)

Equilibrium infection level for the single-strain deterministic model, as a function of the average maturation time. Viral dynamics were simulated under periodic antiviral therapy given by the simple on-off model with a period (T) of 2 days and varying drug efficacy (f). The infection level (heat map color) is measured as the concentration of matur...

Equilibrium infection level for the single-strain deterministic model, as a function of the average maturation time, when the infected cells can die before maturing. Viral dynamics were simulated under periodic antiviral therapy given by the simple on-off model with a period (T) of 2 days and varying drug efficacy (f). The infection level (heat map...

Average equilibrium infection level for the multi-strain competition deterministic model with on/off drug kinetics, when the drug dosage times are normally distributed, as a function of the average maturation time (1/m). The infection level is measured as the concentration of mature infected cells y, under anti-viral therapy with drug efficacy f =...

Time-course of viral fitness under pharmacologic model of periodic drug treatment. Fitness of the viral strain (measured by the basic reproductive ratio R0) fluctuates in response to drug levels approximated by the simple pharmacological model, Eq (7). (A) Perfect adherence to treatment. (B) Imperfect adherence to treatment: each dose is missed wit...

Stochastic competition between viral strains with different life cycle times, for the entire range of drug efficacies. Viral dynamics were simulated under periodic antiviral therapy given by the simple on-off model with a period (T) of 2 days and varying drug efficacy (f). The fixation probability (heat map color) is measured as the fraction of sim...

Stochastic competition between viral strains with different life cycle times, when infected cells can die before maturing. Viral dynamics were simulated under periodic antiviral therapy given by the simple on-off model with a period (T) of 2 days and varying drug efficacy (f). The fixation probability (heat map color) is measured as the fraction of...

Average equilibrium infection level for the single-strain competition deterministic model with pharmacological drug kinetics, as a function of the average maturation time (1/m). The infection level (heat map color) is measured as the concentration of mature infected cells (y) once a steady-state has been reached. Each calculation included only a si...

Competition between viral strains with different life cycle times when there are trade-offs in maturation time. We consider a model in which there is a trade-off to the time spent in the immature phase, even in the absence of drug. The longer the time spent in this phase, the higher the eventual viral burst size and infectivity, but the more chance...

Maturation time of optimal strains across multiple models. The optimal strain is defined as the one with the highest time-average equilibrium level of mature infected cells (y), either when simulated alone (single strain models) or in competition with other strains (multi-strain model). In all cases we only looked at strains with maturation times l...

Stream plot of the vector fields y˙ and x˙ for the infection level y and the healthy cells level x in the single-strain model with n = 2 maturation steps, and constant drug level. (A) Stability of no-infection steady state when R0 < 1. (B) Stability of infection steady state when R0 > 1. The corresponding equations are shown in Eq (1) and Eq. S.1....

Stream plot of the vector fields y˙1,2 and x˙ for the infection levels y1,2 in the 2-strain (N = 2) competition model with constant drug level. (A) Stability of no-infection steady state when R0 < 1 for both viral strains. (B) Stability of infection steady state when R0 > 1 for both viral strains. The corresponding equations are shown in Eq. S.3. T...

Stream plot of the vector fields y˙1,2 and x˙ for the infection levels y1,2 in the 2-strain (N = 2) competition model with constant drug level. (A) Stability of no-infection steady state when R0 < 1 for both viral strains and m1 > m2. (B) Stability of infection steady state when the basic reproductive ratio R0 > 1 for both viral strains and m1 > m2...

[This corrects the article DOI: 10.1371/journal.ppat.1005535.].

Despite antiretroviral therapy (ART), a latent reservoir of replication-competent HIV-1 persists in resting memory CD4 ⁺ T-cells and precludes cure ¹⁻⁶ . Lorenzo-Redondo et al . ⁷ analyzed HIV-1 sequences collected from three individuals during the first six months of ART, discovered specific patterns of sequence evolution, and concluded that viral...

A latent reservoir for HIV-1 in resting CD4 ⁺ T lymphocytes precludes cure. Mechanisms underlying reservoir stability are unclear. Recent studies suggest an unexpected degree of infected cell proliferation in vivo. T cell activation drives proliferation but also reverses latency, resulting in productive infection that generally leads to cell death....

Precision medicine in cancer proposes that genomic characterization of tumors can inform personalized targeted therapies. However, this proposition is complicated by spatial and temporal heterogeneity. Here we study genomic and expression profiles across 127 multisector or longitudinal specimens from 52 individuals with glioblastoma (GBM). Using bu...

Amidst the growing literature on cancer genomics and intratumor heterogeneity, essential principles in evolutionary biology recur time and time again. Here we use these principles to guide the reader through major advances in cancer research, highlighting issues of “hit hard, hit early” treatment strategies, drug resistance, and metastasis. We dist...

A priority for HIV cure research is measuring latent infection that can fuel viral recrudescence if patients cease antiretroviral therapy (ART). One important quantity that is difficult to measure is the rate at which latently infected cells activate, giving rise to spreading infection [1]. Pinkevych et al. recently estimated this rate by analyzing...

Supplementary Methods and Results. The sources of data, model simulation, and parameter estimation are described, and results for fitting models to cohorts 1, 2, and 4 are presented. (PDF)

Glioblastoma (GBM) is the most common and aggressive primary brain tumor. To better understand how GBM evolves, we analyzed longitudinal genomic and transcriptomic data from 114 patients. The analysis shows a highly branched evolutionary pattern in which 63% of patients experience expression-based subtype changes. The branching pattern, together wi...

Monitoring the efficacy of novel reservoir-reducing treatments for HIV is challenging. The limited ability to sample and quantify latent infection means that supervised antiretroviral therapy (ART) interruption studies are generally required. Here we introduce a set of mathematical and statistical modeling tools to aid in the design and interpretat...

Estimating reservoir re-seeding during viral rebound. (PDF)

Estimating the size and source of the reservoir post-transplant. (PDF)

Estimating reservoir size with uncertainty in viral dynamics parameters. (PDF)

Combination antiretroviral therapy (cART) can reduce HIV-1 viremia to clinically undetectable levels. However, replication competent virus persists in a long-lived latent reservoir n resting, memory CD4+ T cells. The latent reservoir in resting CD4+ T cells is the major barrier to curing HIV-1 infection. The recent case of the Berlin patient has su...

Horizontal gene transfers, particularly among microorganisms, complicate efforts to analyze and depict phylogenetic relationships. Topological data analysis provides a novel approach to characterizing both vertical and horizontal modes of evolution that are embedded in genomic datasets. Aligned genomic sequences can be viewed as points in a high-di...

Limiting dilution assays are commonly used to measure the extent of infection, and in the context of HIV they represent an essential tool for studying latency and potential curative strategies. Yet standard assay designs may not discern whether an intervention reduces an already miniscule latent infection. This review addresses challenges arising i...

Significance The evolution of drug resistance is a major health threat. In chronic infections with rapidly mutating pathogens—including HIV, tuberculosis, and hepatitis B and C viruses—multidrug resistance can cause even aggressive combination drug treatment to fail. Oftentimes, individual drugs within a combination do not penetrate equally to all...

Limiting dilution assays are commonly used to measure the extent of infection, and in the context of HIV they represent an essential tool for studying latency and potential curative strategies. To assist investigators using dilution assays, we illustrate the major statistical method for estimating the frequency of infected cells (or other infectiou...

Reversal of HIV-1 latency by small molecules is a potential cure strategy. This approach will likely require effective drug combinations to achieve high levels of latency reversal. Using resting CD4+ T cells (rCD4s) from infected individuals, we developed an experimental and theoretical framework to identify effective latency-reversing agent (LRA)...

Recently the “Mississippi Child” was reported as having a prolonged clearance of HIV viremia after the initiation of antiretroviral therapy shortly after birth. Further follow-up of this case is now reported.

Infections with rapidly evolving pathogens are often treated using combinations of drugs with different mechanisms of action. One of the major goals of combination therapy is to reduce the risk of drug resistance emerging during a patient's treatment. While this strategy generally has significant benefits over monotherapy, it may also select for mu...

The viral reservoir represents a critical challenge for human immunodeficiency virus type 1 (HIV-1) eradication strategies. However, it remains unclear when and where the viral reservoir is seeded during acute infection and the extent to which it is susceptible to early antiretroviral therapy (ART). Here we show that the viral reservoir is seeded r...

Persistent homology computes topological invariants from point cloud data. Recent work has focused on developing statistical methods for data analysis in this framework. We show that, in certain models, parametric inference can be performed using statistics defined on the computed invariants. We develop this idea with a model from population geneti...

Abstract Theoretical and experimental studies have shown that high mutation rates can be advantageous, especially in novel or fluctuating environments. Here we examine how frequency-dependent competition may lead to fluctuations in trait frequencies that exert upward selective pressure on mutation rates. We use a mathematical model to show that cyc...

Massive research efforts are now underway to develop a cure for HIV infection, allowing patients to discontinue lifelong combination antiretroviral therapy (ART). New latency-reversing agents (LRAs) may be able to purge the persistent reservoir of latent virus in resting memory CD4+ T cells, but the degree of reservoir reduction needed for cure rem...

Antiretroviral therapy fails to cure HIV-1 infection because latent proviruses persist in resting CD4(+) T cells. T cell activation reverses latency, but <1% of proviruses are induced to release infectious virus after maximum in vitro activation. The noninduced proviruses are generally considered defective but have not been characterized. Analysis...

Despite the high inhibition of viral replication achieved by current anti-HIV drugs, many patients fail treatment, often with emergence of drug-resistant virus. Clinical observations show that the relationship between adherence and likelihood of resistance differs dramatically among drug classes. We developed a mathematical model that explains thes...

The mutation rate of an organism is itself evolvable. In stable environments, if faithful replication is costless, theory predicts that mutation rates will evolve to zero. However, positive mutation rates can evolve in novel or fluctuating environments, as analytical and empirical studies have shown. Previous work on this question has focused on en...

Infectious diseases remain a formidable challenge to human health, and understanding pathogen evolution is crucial to designing effective therapeutics and control strategies. Here, we review important evolutionary aspects of HIV infection, highlighting the concept of selection at multiple spatial and temporal scales. At the smallest scale, a single...

The problem of achieving widespread immunity to infectious diseases by voluntary vaccination is often presented as a public-goods dilemma, as an individual's vaccination contributes to herd immunity, protecting those who forgo vaccination. The temptation to free-ride brings the equilibrium vaccination level below the social optimum. Here, we presen...