<title>Author Summary</title> <p>Binding of transcription factors to promoters of genes, and subsequent enhancement or repression of transcription, is one of the main steps of transcriptional gene regulation. Direct or indirect wet-lab experiments allow the identification of approximate regions potentially bound or regulated by a transcription factor. Subsequently, de-novo motif discovery tools can be used for detecting the precise positions of binding sites. Many traditional tools focus on motifs over-represented in the target regions, which often turn out to be similarly over-represented in the entire genome. In contrast, several recent tools focus on differentially abundant motifs in target regions compared to a control set. As binding sites are often located at some preferred distance to the transcription start site, it is favorable to include this information into de-novo motif discovery. Here, we present Dispom a novel approach for learning differentially abundant motifs and their positional preferences simultaneously, which predicts binding sites with increased accuracy compared to many popular de-novo motif discovery tools. When applying Dispom to promoters of auxin-responsive genes of Arabidopsis thaliana, we find a binding motif slightly different from the canonical auxin-response element, which exhibits a strong positional preference and which is considerably more specific to auxin-responsive genes.</p>

Keywords: automatic documentation, meta data, XML, processing graph, image analysis

BACKGROUND:The recognition of functional binding sites in genomic DNA remains one of the fundamental challenges of genome research. During the last decades, a plethora of different and well-adapted models has been developed, but only little attention has be payed to the development of different and similarly well-adapted learning principles. Only recently it was noticed that discriminative learning principles can be superior over generative ones in diverse bioinformatics applications, too.RESULTS:Here, we propose a generalization of generative and discriminative learning principles containing the maximum likelihood, maximum a-posteriori, maximum conditional likelihood, maximum supervised posterior, generative-discriminative trade-off, and penalized generative-discriminative trade-off learning principles as special cases, and we illustrate its efficacy for the recognition of vertebrate transcription factor binding sites.CONCLUSIONS:We find that the proposed learning principle helps to improve the recognition of transcription factor binding sites and enables better computational approaches for extracting as much information as possible from valuable wet-lab data. We make all implementations available in the open-source library Jstacs so that this learning principle can be easily applied to other classification problems in the field of genome and epigenome analysis.

BACKGROUND: One of the challenges of bioinformatics remains the recognition of short signal sequences in genomic DNA such as donor or acceptor splice sites, splicing enhancers or silencers, translation initiation sites, transcription start sites, transcription factor binding sites, nucleosome binding sites, miRNA binding sites, or insulator binding sites. During the last decade, a wealth of algorithms for the recognition of such DNA sequences has been developed and compared with the goal of improving their performance and to deepen our understanding of the underlying cellular processes. Most of these algorithms are based on statistical models belonging to the family of Markov random fields such as position weight matrix models, weight array matrix models, Markov models of higher order, or moral Bayesian networks. While in many comparative studies different learning principles or different statistical models have been compared, the influence of choosing different prior distributions for the model parameters when using different learning principles has been overlooked, and possibly lead to questionable conclusions. RESULTS: With the goal of allowing direct comparisons of different learning principles for models from the family of Markov random fields based on the same a-priori information, we derive a generalization of the commonly-used product-Dirichlet prior. We find that the derived prior behaves like a Gaussian prior close to the maximum and like a Laplace prior in the far tails. In two case studies, we illustrate the utility of the derived prior for a direct comparison of different learning principles with different models for the recognition of binding sites of the transcription factor Sp1 and human donor splice sites. CONCLUSIONS: We find that comparisons of different learning principles using the same a-priori information can lead to conclusions different from those of previous studies in which the effect resulting from different priors has been neglected. We implement the derived prior is implemented in the open-source library Jstacs to enable an easy application to comparative studies of different learning principles in the field of sequence analysis.