Systems biology has matured considerably as a discipline over the last decade, yet some of the key challenges separating current research efforts in systems biology and clinically useful results are only now becoming apparent. As these gaps are better defined, the new discipline of systems medicine is emerging as a translational extension of systems biology. How is systems medicine defined? What are relevant ontologies for systems medicine? What are the key theoretic and methodologic challenges facing computational disease modeling? How are inaccurate and incomplete data, and uncertain biologic knowledge best synthesized in useful computational models? Does network analysis provide clinically useful insight? We discuss the outstanding difficulties in translating a rapidly growing body of data into knowledge usable at the bedside. Although core-specific challenges are best met by specialized groups, it appears fundamental that such efforts should be guided by a roadmap for systems medicine drafted by a coalition of scientists from the clinical, experimental, computational, and theoretic domains.
17 July 2019 in Proceedings of the AAAI Conference on Artificial Intelligence
We develop a non-convex optimization approach to correlation clustering using the Frank-Wolfe (FW) framework. We show that the basic approach leads to a simple and natural local search algorithm with guaranteed convergence. This algorithm already beats alternative algorithms by substantial margins in both running time and quality of the clustering. Using ideas from FW algorithms, we develop subsampling and variance reduction paradigms for this approach. This yields both a practical improvement of the algorithm and some interesting further directions to investigate. We demonstrate the performance on both synthetic and real world data sets.
Background: Temporal analysis of gene expression data has been limited to identifying genes whose expression varies with time and/or correlation between genes that have similar temporal profiles. Often, the methods do not consider the underlying network constraints that connect the genes. It is becoming increasingly evident that interactions change substantially with time. Thus far, there is no systematic method to relate the temporal changes in gene expression to the dynamics of interactions between them. Information on interaction dynamics would open up possibilities for discovering new mechanisms of regulation by providing valuable insight into identifying time-sensitive interactions as well as permit studies on the effect of a genetic perturbation. Results: We present NETGEM, a tractable model rooted in Markov dynamics, for analyzing the dynamics of the interactions between proteins based on the dynamics of the expression changes of the genes that encode them. The model treats the interaction strengths as random variables which are modulated by suitable priors. This approach is necessitated by the extremely small sample size of the datasets, relative to the number of interactions. The model is amenable to a linear time algorithm for efficient inference. Using temporal gene expression data, NETGEM was successful in identifying (i) temporal interactions and determining their strength, (ii) functional categories of the actively interacting partners and (iii) dynamics of interactions in perturbed networks. Conclusions: NETGEM represents an optimal trade-off between model complexity and data requirement. It was able to deduce actively interacting genes and functional categories from temporal gene expression data. It permits inference by incorporating the information available in perturbed networks. Given that the inputs to NETGEM are only the network and the temporal variation of the nodes, this algorithm promises to have widespread applications, beyond biological systems. The source code for NETGEM is available from https://github.com/vjethava/NETGEM
In Leo Breiman's influential article "Statistical modeling-the two cultures" he identified two cultures for statistical practices. The data modeling culture (DMC) denotes practices tailored for statistical inference targeting a quantity of interest, [inline-graphic 01]. The algorithmic modeling culture (AMC) refers to practices defining an algorithm, or a machine-learning (ML) procedure, that generates accurate predictions about an outcome of interest, [inline-graphic 02] was the dominant mode, Breiman argued that statisticians should give more attention to AMC. Twenty years later and energized by two revolutions—one in data-science and one in causal inference—a hybrid modeling culture (HMC) is rising. HMC fuses the inferential strength of DMC and the predictive power of AMC with the goal of analyzing cause and effect, and thus, HMC's quantity of interest is causal effect, [inline-graphic 03]. In combining inference and prediction, the result of HMC practices is that the distinction between prediction and inference, taken to its limit, melts away. While this hybrid culture does not occupy the default mode of scientific practices, we argue that it offers an intriguing novel path for applied sciences.
Recent results on optimization and generalization properties of neural networks showed that in a simple two-layer network, the alignment of the labels to the eigenvectors of the corresponding Gram matrix determines the convergence of the optimization during training. Such analyses also provide upper bounds on the generalization error. We experimentally investigate the implications of these results to deeper networks via embeddings. We regard the layers preceding the final hidden layer as producing different representations of the input data which are then fed to the two-layer model. We show that these representations improve both optimization and generalization. In particular, we investigate three kernel representations when fed to the final hidden layer: the Gaussian kernel and its approximation by random Fourier features, kernels designed to imitate representations produced by neural networks and finally an optimal kernel designed to align the data with target labels. The approximated representations induced by these kernels are fed to the neural network and the optimization and generalization properties of the final model are evaluated and compared.