Web supplement to "A compendium of nucleosome profiles."

Main figures and tables (in no particular order for now)

• Figure 1, Regulators and coregulators. pdf - illustrator - treeview
• Figure 2, CUTs at NFRs for regulator mutants. pdf - illustrator
• Figure 3, Averagograms to highlight major defects and spacing changes. pdf - illustrator
• Figure 4, Highlight specific changes in Ino80 and Chd1 mutants (affected genes; compare with Isw1?)
• Figure 5, Tup1/Cyc8 changes
• Figure 6, Antisense transcripts in mutants with disrupted nucleosomes in gene bodies (start site variability)
• Figure 7, NFR/expression changes for all other categories (Remodelers, Modifiers, RNAPI/II, Histone depletion)

Supplementary figures and tables

• Supplementary table 1, Strain information. excel
• Supplementary figure 1, Correlation between tiling arary replicates
• Supplementary figure 2, MCM1 NFR cuts and absence of NFR cuts upon histone depletion (controls for figure 2)
• Supplementary figure 3, Full version of Averagogram figure (with all names visible)
• Supplementary figure 4, Full version of the Tup1/Cyc8 changes

Supplementary genome browser

• Gbrowse supplementary site

• Prepare the Tup1/Cyc8 figures. bar files - matrices - matrices-3kb-flanking - target genes - "orphan" nucleosome changes (Kyle)
1. Curate Tup1/Cyc8 derepressed genes
2. Supplementary figure showing full changes across all genes (sorted by Tup1/Cyc8 NFR change)
3. Panel for figure 5, highlighting scope of changes across all target genes (cluster diagram)
4. Panel for figure 5 with an example region
5. Supplementary table with Tup1/Cyc8 target genes
6. Catalog the new transcripts found in the telomeric and other regions in Tup1 and Cyc8 mutants
• Identify genes with nucleosome shifts in genes in the Ino80 and Chd1 mutants for figure 4 (Harm, need seq data)
• Prepare figure 6 on the antisense transcripts originating in ORFs. (Harm)
• Related to Story 1, can we demonstrate that the new transcripts are indeed CUTs? Northern?
• Related to Story 1, why do some NFR changes result in CUTs, but not others? Occupancy vs. position change? (Harm, need seq data)
• Set up genome browser with all nuc/transcription data (Harm, new server ordered)

Bar files
The bar files contain the normalized probe intensity data for each experiment. They are best viewed in the integrated genome browser (IGB). After starting IGB, go to the "Data access" tab and select "Saccharomyces cerevisiae" as the organism and "S_cerevisiae_May_2008" as the genome version. Under "Choose data sources and data sets", there should now be an entry labeled "Hugheslab (quickload)". Click this entry to show the available data tracks and select "sgd_orfs" (and any other track of interest). The ORF annotations should now appear in the main window. To view the bar file tracks, go to the "File" menu and open the files that you want to see.

• Nucleosome data, normalized to genomic DNA (2.2 Gb)
• Nucleosome data, changes compared to wild-type (1.6 Gb)
• Expression data, changes compared to wild-type (1.2 Gb)

Clustering data for each strain

• Gene data (7.7 Gb)
• Non-gene data (7.7 Gb)

The links above contain the combined clustering diagrams for the nucleosome and transcription data for gene and non-gene features. For each mutant, the data is further subdivided into sense and antisense transcription data, with clustering relative to the transcription start or termination site. See also this pdf file for more info on how to interpret the diagrams.

Here is the breakdown of data for gene features:

• combined_gene/sense-tss/ : Nucleosome data relative to the TSS; Sense transcription data
• combined_gene/sense-tts/ : Nucleosome data relative to the TTS; Sense transcription data
• combined_gene/antisense-tss/ : Nucleosome data relative to the TSS; Antisense transcription data
• combined_gene/antisense-tts/ : Nucleosome data relative to the TTS; Antisense transcription data

Within each of these folders, there are five types of data:

• Weighted : Pearson clustering on NFR and transcription changes
• nucsort : Data sorted by change in NFR occupancy
• genesort : Data sorted by gene transcription change
• 5primesort : Data sorted by transcription change at 5' end of gene
• 3primesort : Data sorted by transcription change at 3' end of gene

The sorted clustering diagrams are useful to quickly assess if there is correlation between nucleosome occupancy and transcription. Each data type has an associated '.cdt' file that can be opened with Java Treeview and contains annotation details and GO category information that can be displayed alongside the cluster diagrams. For each clustering I also auto-generated a '.png' preview file, so that you can check the clustering for major effects, before going through the effort of opening all cluster diagrams.

Below is the breakdown for the non-gene features. Note that I did not distinguish between the TSS or TTS for the non-gene features (all is relative to the TSS) since they are typically so small that the results would essentially be the same.

• ARS/ : Data relative to origins of replication
• centromere/ : Data relative to centromeres
• LTR_retrotransposon/ : Data relative to LTRs
• snoRNAs/ : Data relative to snoRNAs
• tRNA/ : Data relative to tRNAs

Within each folder there are two types of data:

• unweighted : Pearson clustering of nuc & transcription data
• nucsort : Data sorted by change in NFR occupancy