The Chromatin Organization of an Eukaryotic Genome : Sequence Specific+ Statistical=Combinatorial (Extended Abstract)

Nucleosome organization in eukaryotic genomes has a deep impact on gene function. Although progress has been recently made in the identification of various concurring factors influencing nucleosome positioning, it is still unclear whether nucleosome positions are sequence dictated or determined by a random process. It has been postulated for a long time that,in the proximity of TSS, a barrier determines the position of the +1 nucleosome and then geometric constraints alter the random positioning process determining nucleosomal phasing. Such a pattern fades out as one moves away from the barrier to become again a random positioning process. Although this statistical model is widely accepted,the molecular nature of the barrier is still unknown. Moreover,we are far from the identification of a set of sequence rules able:to account for the genome-wide nucleosome organization;to explain the nature of the barriers on which the statistical mechanism hinges;to allow for a smooth transition from sequence-dictated to statistical positioning and back. We show that sequence complexity,quantified via various methods, can be the rule able to at least partially account for all the above.In particular, we have conducted our analyses on 4 high resolution nucleosomal maps of the model eukaryotes and found that nucleosome depleted regions can be well distinguished from nucleosome enriched regions by sequence complexity measures.In particular, (a) the depleted regions are less complex than the enriched ones, (b) around TSS complexity measures alone are in striking agreement with in vivo nucleosome occupancy,in particular precisely indicating the positions of the +1 and -1 nucleosomes. Those findings indicate that the intrinsic richness of subsequences within sequences plays a role in nucleosomal formation in genomes, and that sequence complexity constitutes the molecular nature of nucleosome barrier.

💡 Research Summary

The paper tackles the long‑standing debate over whether nucleosome positioning across eukaryotic genomes is dictated primarily by DNA sequence or by a stochastic, statistical process. Classical models propose that a “barrier” near transcription start sites (TSS) fixes the +1 nucleosome, after which geometric constraints generate phased nucleosome arrays that gradually dissolve into random placement further away. However, the molecular identity of this barrier has remained elusive, and a unified rule that can explain both sequence‑driven and statistical positioning has been lacking.

The authors introduce sequence complexity as a quantitative descriptor of intrinsic DNA information content. Using several established metrics—including Lempel‑Ziv compression, Shannon entropy, and k‑mer diversity—they compute complexity scores for sliding windows across four high‑resolution nucleosome maps derived from yeast, Drosophila, and human cells. By correlating these scores with experimentally measured nucleosome occupancy, they uncover a striking pattern: nucleosome‑depleted regions (NDRs) exhibit significantly lower complexity than nucleosome‑enriched regions. Moreover, the complexity profile around TSS aligns precisely with the positions of the –1 and +1 nucleosomes, effectively reproducing the “barrier” that classical statistical models require.

These findings suggest that low‑complexity DNA sequences act as physical or energetic barriers that impede nucleosome formation, while high‑complexity sequences provide a favorable substrate for nucleosome assembly. In this view, the statistical positioning mechanism operates downstream of a sequence‑defined landscape: nucleosomes are first excluded from low‑complexity zones, then become phased by steric constraints, and finally revert to a more random distribution as complexity levels off. The study therefore bridges the gap between purely sequence‑based and purely statistical explanations, offering a unified combinatorial framework.

Beyond the immediate mechanistic insight, the authors propose that complexity measures could be integrated with other epigenetic signals—such as transcription‑factor binding motifs, histone modifications, and evolutionary conservation—to improve genome‑wide predictions of chromatin architecture. By identifying sequence complexity as the molecular nature of the nucleosome barrier, the work opens new avenues for computational modeling of chromatin organization, rational design of synthetic promoters, and targeted manipulation of nucleosome positioning in therapeutic contexts.