methylFlow: cell-specific methylation pattern reconstruction from high-throughput bisulfite-converted DNA sequencing.

MOTIVATION: DNA methylation aberrations are now known to, almost universally, accompany the initiation and progression of cancers. In particular, the colon cancer epigenome contains specific genomic regions that, along with differences in methylation levels with respect to normal colon tissue, also show increased epigenetic and gene expression heterogeneity at the population level, i.e. across tumor samples, in comparison with other regions in the genome. Tumors are highly heterogeneous at the clonal level as well, and the relationship between clonal and population heterogeneity is poorly understood. RESULTS: We present an approach that uses sequencing reads from high-throughput sequencing of bisulfite-converted DNA to reconstruct heterogeneous cell populations by assembling cell-specific methylation patterns. Our methodology is based on the solution of a specific class of minimum cost network flow problems. We use our methods to analyze the relationship between clonal heterogeneity and population heterogeneity in high-coverage data from multiple samples of colon tumor and matched normal tissues. AVAILABILITY AND IMPLEMENTATION: http://github.com/hcorrada/methylFlow CONTACT: hcorrada@umiacs.umd.edu SUPPLEMENTARY INFORMATION: SUPPLEMENTARY INFORMATION is available at Bioinformatics online.

Stochastic Cell Fate and Longevity of Offspring.

OBJECTIVE: Cellular decision-making is a key process in which cells with similar geneticand environmental background make dissimilar decisions. This stochastic process, which happens in prokaryotic and eukaryotic cells including stem cells, causes cellular diversity and phenotypic variation. In addition, fitness predicts and describes changes in the genetic composition of populations throughout the evolutionary history. Fitness may thus be defined as the ability to adapt and produce surviving offspring. Here, we present a mathematical model to predict the fitness of a cell and to address the fundamental issue of phenotypic variation. We study a basic decision-making scenario where a bacteriophage lambda reproduces in E. coli, using both the lytic and the lysogenic pathways. In the lytic pathway, the bacteriophage replicates itself within the host bacterium. This fast replication overcrowds and in turn destroys the host bacterium. In the lysogenic pathway, however, the bacteriophage inserts its DNA into the host genome, and is replicated simultaneously with the host genome. MATERIALS AND METHODS: In this prospective study, a mathematical predictive model was developed to estimate fitness as an index of survived offspring. We then leverage experimental data to validate the predictive power of our proposed model. A mathematical model based on game theory was also generated to elucidate a rationale behind cell decision. RESULTS: Our findings indicate that a rational decision that is aimed to maximize life expectancy of offspring is almost identical to bacteriophage behavior reported based on experimental data. The results also showed that stochastic decision on cell fate maximizes the expected number of survived offspring. CONCLUSION: We present a mathematical framework for analyzing a basic phenotypic variation problem and explain how bacteriophages maximize offspring longevity based on this model. We also introduce a mathematical benchmark for other investigations of phenotypic variation that exists in eukaryotes including stem cell differentiation.

Natural biased coin encoded in the genome determines cell strategy.

Decision making at a cellular level determines different fates for isogenic cells. However, it is not yet clear how rational decisions are encoded in the genome, how they are transmitted to their offspring, and whether they evolve and become optimized throughout generations. In this paper, we use a game theoretic approach to explain how rational decisions are made in the presence of cooperators and competitors. Our results suggest the existence of an internal switch that operates as a biased coin. The biased coin is, in fact, a biochemical bistable network of interacting genes that can flip to one of its stable states in response to different environmental stimuli. We present a framework to describe how the positions of attractors in such a gene regulatory network correspond to the behavior of a rational player in a competing environment. We evaluate our model by considering lysis/lysogeny decision making of bacteriophage lambda in E. coli.

Missing value imputation in DNA microarrays based on conjugate gradient method.

Analysis of gene expression profiles needs a complete matrix of gene array values; consequently, imputation methods have been suggested. In this paper, an algorithm that is based on conjugate gradient (CG) method is proposed to estimate missing values. k-nearest neighbors of the missed entry are first selected based on absolute values of their Pearson correlation coefficient. Then a subset of genes among the k-nearest neighbors is labeled as the best similar ones. CG algorithm with this subset as its input is then used to estimate the missing values. Our proposed CG based algorithm (CGimpute) is evaluated on different data sets. The results are compared with sequential local least squares (SLLSimpute), Bayesian principle component analysis (BPCAimpute), local least squares imputation (LLSimpute), iterated local least squares imputation (ILLSimpute) and adaptive k-nearest neighbors imputation (KNNKimpute) methods. The average of normalized root mean squares error (NRMSE) and relative NRMSE in different data sets with various missing rates shows CGimpute outperforms other methods.