Browsing by Author "Schultz, Andre"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    A Crowdsourcing Approach to Developing and Assessing Prediction Algorithms for AML Prognosis
    (Public Library of Science, 2016) Noren, David P.; Long, Byron L.; Norel, Raquel; Rrhissorrakrai, Kahn; Hess, Kenneth; Hu, Chenyue Wendy; Bisberg, Alex J.; Schultz, Andre; Engquist, Erik; Liu, Li; Lin, Xihui; Chen, Gregory M.; Xie, Honglei; Hunter, Geoffrey A.M.; Boutros, Paul C.; Stepanov, Oleg; DREAM 9 AML-OPC Consortium; Norman, Thea; Friend, Stephen H.; Stolovitzky, Gustavo; Kornblau, Steven; Qutub, Amina A.; Bioengineering
    Acute Myeloid Leukemia (AML) is a fatal hematological cancer. The genetic abnormalities underlying AML are extremely heterogeneous among patients, making prognosis and treatment selection very difficult. While clinical proteomics data has the potential to improve prognosis accuracy, thus far, the quantitative means to do so have yet to be developed. Here we report the results and insights gained from the DREAM 9 Acute Myeloid Prediction Outcome Prediction Challenge (AML-OPC), a crowdsourcing effort designed to promote the development of quantitative methods for AML prognosis prediction. We identify the most accurate and robust models in predicting patient response to therapy, remission duration, and overall survival. We further investigate patient response to therapy, a clinically actionable prediction, and find that patients that are classified as resistant to therapy are harder to predict than responsive patients across the 31 models submitted to the challenge. The top two performing models, which held a high sensitivity to these patients, substantially utilized the proteomics data to make predictions. Using these models, we also identify which signaling proteins were useful in predicting patient therapeutic response.
  • Thumbnail Image
    Item
    Improved Methods for Constraint Based Modeling of Mammalian Systems
    (2017-06-26) Schultz, Andre; Qutub, Amina A
    Genome-wide metabolic reconstructions have been widely applied to study metabolism at a genome scale. To date, most of the work in the field has been performed in the study of unicellular organisms, however, and many of the methods developed in this context do not transfer for the study of mammalian systems. For instance, (1) the larger size of mammalian reconstructions makes the application of computationally expensive algorithms such as pathway decomposition infeasible. Also, (2) the optimization of a cellular objective, commonly defined to be biomass production in unicellular organisms, does not transfer to mammalian cells, where a cellular objective is neither well defined nor optimized. Finally, (3) the generalized human reconstruction needs to be tailored to specific tissues or cell lines for a context specific analysis, since only a subset of the metabolism defined in the human genome takes place in each cell. In this project, we aim to develop better methods for the analysis of mammalian systems using genome-scale models. We demonstrate that (1) the removal of currency metabolites and energy related loops from the model leads to a more feasible and biologically relevant application of pathway decomposition analysis. We also show that merging sets of fully coupled reactions, and using a combination of two algorithms, leads to a significantly faster implementation of Monte-Carlo sampling. Furthermore, (2) by fixing the cellular objective and optimizing metabolic resources, we demonstrate that a sub-optimal objective oriented approach can significantly improve flux prediction results. Finally, (3) we present a context-specific algorithm that is faster, agrees better with experimental data, and yields better tissue-specific predictions when compared to previous methods. After validating these methods, we apply them to the study of E. coli metabolism, cancer cells in a subtype specific manner, and to the study of hypoxia adaptation in mouse cardiomyocytes. Results from our predictions provided biological insight in both applications: including the role of Hexosamine synthesis pathways as an energy regulator in cancer cells, and the role of evolution in the adaptation of mouse populations to altitude conditions.