General Biology Seminar

Abstract: How transcriptional networks direct cell-fate specification in a reliable and reproducible manner is an important question in developmental biology. Drosophila segmentation has emerged as a key model for investigating how gene expression patterns are specified reliably. I will present evidence that an enhancer of the segmentation gene even-skipped contains sequences that are dispensable both for qualitatively recapitulating endogenous gene expression and for adult viability in rescue experiments. These sequences are however required for precise placement of gene expression domains and for temperature compensation. Next I will show that even-skipped has a sexually dimorphic expression pattern, which can be traced to the incomplete dosage compensation of an X-linked gene, giant. Segmentation itself is not dimorphic, implying that the segmentation gene network corrects the perturbation created by incomplete dosage compensation. In the second part of the talk I will describe gene regulation during hematopoiesis, another developmental system largely controlled by transcriptional networks. Gene expression is controlled by multiple cis-regulatory modules (CRMs), each consisting of several transcription factor binding sites. Inferring the logic of cis regulation remains a low-throughput process despite the availability of genomic data. I will describe a strategy that uses a sequence-based model of transcription and quantitative gene expression data to speed up the reverse engineering process. Using this methodology, I will show that C/EBPα, a myeloid gene, has a complex regulatory logic that integrates multiple myeloid and non-myeloid inputs.