Nucleosome collections

Note: Di-nucleotide frequencies below are computed from the center alignment, using a 3 basepair moving average window. Yeast nucleosome-bound sequences were measured by us, chicken nucleosome-bound sequences were measured by Satchwell et al. (JMB, 1986), and the in vitro selected nucleosome-bound sequences were measured by Lowary et al. (JMB, 1998).

	Yeast (in vivo)	Yeast (in vitro)	Chicken (in vivo)	Mouse (in vitro)	In Vitro selection
Sequences
Center alignment
Di-nucleotide frequencies (center alignment)

In vivo nucleosome occupancy measurements at regions with high predicted occupancy

Experimental data (details are explained in the spreadsheet. See text for additional details).

In vitro nucleosome binding affinity measurements of mutated sequences

	Tested sequences (aligned as in Figure)
	Binding affinities to all tested sequences

Genome-wide encoded nucleosome organization of Yeast (predicted by our model)

Files below provide details of our predictions, including the set of predicted highly stable nucleosomes, the predicted nucleosome occupancy at each basepair, the single best nucleosome organization on the genome, and the raw nucleosome binding energy landscape per basepair, which is the basis of the genome-wide organizations. Files have six columns with the following format: chromosome (column 1), unique id (column 2), start basepair (column 3), end basepair (column 4), internal use (column 5), value (column 6).

	Yeast model predictions	Chicken model predictions
Highly stable nucleosomes (stability probability > 0.5)
Highly stable nucleosomes (stability probability > 0.2)
Nucleosome occupancy at each basepair in yeast genome
Single best genome-wide nucleosome configuration
Raw nucleosome binding log-ratio per basepair
Probability of starting a nucleosome at each basepair

Since the publication of our paper, we have been working on improving our model in several directions. Below we provide predictions from our current working model, updated March 2008:

	Updated model
Nucleosome occupancy at each basepair in yeast genome
Raw nucleosome binding log-ratio per basepair in yeast genome

Encoded nucleosome occupancy by chromosomal regions (predicted by our model)

For each type of chromosomal region below, the associated table below provides the average nucleosome occupancy predicted by our model across each genomic region of the associated type (for example, the 'coding region' table provides the average nucleosome occupancy per basepair predicted by our model for the coding region of each of gene in yeast).

	Yeast model predictions
Coding regions
Intergenic (unique promoter)
Intergenic (divergent promoter)
Intergenic (not a promoter)
Centromeres
Autonomously Replicating Sequences (ARS)
Telomeres
Protein-DNA binding sites
Ribosomal RNAs
Transfer RNAs (tRNAs)

In-vivo mapped nucleosome positions

Genomic locations of in-vivo nucleosomes.

Literature-mapped nucleosome positions

Literature mapped nucleosomes (99 nucleosomes at 11 loci).

Yeast Genome

The data files above were generated and are given in coordinates that correspond to the yeast genome version downloaded from SGD on January 2006, and is available here.