Stochastic pairing of docking sequence with any one of the selector sequences 5' of the mutually exclusive exons of Dscam was suggested as one possible mechanism to explain this puzzling gene.
However stochastic selection explains how any single molecule of RNA in neurons can be processed to include a single exon choice, but does not help in understanding how the "memory" of this process is maintained: in fact, any single neuron will express only one single splicing isoforms, and it is difficult to conceive that only one single molecule of Dscam mRNA is synthesized in neuron lifespan.
Recent data show that nucleosomes are preferentially positioned over the exons, with particular frequence over alternative exons, where tehy also show pecific histone modifications.
Could you formulate some hypothesis on Dscam exon selection based on nucleosomal positioning?