Learn More
Semi-Markov Decision Problems are continuous time generalizations of discrete time Markov Decision Problems. A number of reinforcement learning algorithms have been developed recently for the solution of Markov Decision Problems, based on the ideas of asynchronous dynamic programming and stochastic approximation. Among these are TD(), Q-learning, and(More)
Compelling evidence indicates that the CRISPR-Cas system protects prokaryotes from viruses and other potential genome invaders. This adaptive prokaryotic immune system arises from the clustered regularly interspaced short palindromic repeats (CRISPRs) found in prokaryotic genomes, which harbor short invader-derived sequences, and the CRISPR-associated (Cas)(More)
The regulation of gene expression is critical for organismal function and is an important source of phenotypic diversity between species. Understanding the genetic and molecular mechanisms responsible for regulatory divergence is therefore expected to provide insight into evolutionary change. Using deep sequencing, we quantified total and allele-specific(More)
To gain insight into how genomic information is translated into cellular and developmental programs, the Drosophila model organism Encyclopedia of DNA Elements (modENCODE) project is comprehensively mapping transcripts, histone modifications, chromosomal proteins, transcription factors, replication proteins and intermediates, and nucleosome properties(More)
  • James B. Brown, Nathan Boley, Robert Eisman, Gemma E. May, Marcus H. Stoiber, Michael O. Duff +35 others
  • 2014
Animal transcriptomes are dynamic, with each cell type, tissue and organ system expressing an ensemble of transcript isoforms that give rise to substantial diversity. Here we have identified new genes, transcripts and proteins using poly(A)+ RNA sequencing from Drosophila melanogaster in cultured cell lines, dissected organ systems and under environmental(More)
Given a Markov decision process (MDP) with expressed prior uncertainties in the process transition probabilities, we consider the problem of computing a policy that optimizes expected total (finite-horizon) reward. Implicitly, such a policy would effectively resolve the "exploration-versus-exploitation tradeoff" faced, for example, by an agent that seeks to(More)
Accurate gene model annotation of reference genomes is critical for making them useful. The modENCODE project has improved the D. melanogaster genome annotation by using deep and diverse high-throughput data. Since transcriptional activity that has been evolutionarily conserved is likely to have an advantageous function, we have performed large-scale(More)
Substantial evidence suggests that the phasic activities of dopaminergic neurons in the primate midbrain represent a temporal difference (TD) error in predictions of future reward, with increases above and decreases below baseline consequent on positive and negative prediction errors, respectively. However, dopamine cells have very low baseline activity,(More)
BACKGROUND RNAs can be physically classified into poly(A)+ or poly(A)- transcripts according to the presence or absence of a poly(A) tail at their 3' ends. Current deep sequencing approaches largely depend on the enrichment of transcripts with a poly(A) tail, and therefore offer little insight into the nature and expression of transcripts that lack poly(A)(More)