MethylViewer initial window. The initial window after loading MethylViewer.exe (A) and Analysis > Alignment options (B). Minimum alignment score … can be set with increased stringency from 15 to 50 upon activating the pop-up window. Likewise, users can set the Base calling cut off … for the signal threshold (15 to 300) for calling bases in sequencing traces and Word size … of the base pair length of the array of overlapping DNA fragments (6 to 15) created from the wild-type reference sequence for use in the BLAST-like local extension algorithm. Clicking the mouse on Interactive view … or Import FASTA alignment … activates the Methylation sites window shown in Figure 3A.

Default settings for MAPit methylation footprinting analysis. (A) Methylation sites window activated by clicking on Analysis > Interactive view … or Import FASTA alignment … . If the default setting of CG and GC (see keys in Figures 4A and 10D) is acceptable, the user then selects the indicated files as prompted in the window. (B) Custom methylation sites window activated by clicking the Custom button in A. Shown are the default settings for CG methylation analysis: methylated dC residues at position 1 indicated in black. Note that the Include the site field indicates CG: 1. Toggling to GC: 2 shows the second setting, methylation of dC residues at position 2 in GC sites to be indicated in red. The circle and triangle symbols are gray because they are set as defaults for indicating methylation status of CG and GC sites, respectively, in exported images (e.g. Figure 10).

Data editing. (A) Right-clicking on the light blue header cell offers the option to view the entire sequencing trace of a molecule. (B) Right-clicking on any non-header cell in the grid displays (C) the local alignment between the reference and query sequences as well as the sequencing trace encompassing the query site. The methylatable residue in both the reference and query sequences is indicated in red type in properly aligned sequences. In this example, the A residue in the upper query sequence (manually highlighted by red circle; reverse strand sequenced) is in black type, because it should have been aligned to the G in the reference sequence. That is, the gap in the alignment (indicated in A and B as a blue tracer line within the corresponding cell) should be placed at one of the three immediately downstream C residues. Therefore, clicking anywhere in the white field of this local alignment window displays a pop-up window to correct assignment of the nucleotide sequence; specifically, G‘T’, because the corresponding nucleotide in the query molecule sequenced as ‘A’ (when aligned correctly). The base calling cutoff of 20 is indicated by the red horizontal line at the base of the peaks. (D) The corrected G‘T’ appears as a large white square within the original yellow cell for that nucleotide. Cells for which sequencing data was inspected and verified, and hence had unchanged sequencing calls, are marked by small green squares. The window scroll bar has been moved all the way to the right in this panel. (E) Right-clicking on the CG cell marked by the cursor in D showed deletion of an A in the query molecule. As this A or any other in the downstream run of adenines could have been deleted, the nucleotide assignment of the residue was changed to Not aligned. If FASTA files are used as input, right clicking on any non-header cell in the grid will display the entire text alignment between the reference and query sequence, with the inspected nucleotide marked by an asterisk.

Edited interactive grids for mammalian MAPit data. Nuclei from cell line HCT116 + 0 U M.CviPI (A), HCT116 + 100 U M.CviPI (B), RKO + 0 U M.CviPI (C) and RKO + 100 U M.CviPI (D). The same *.ab1 files used in Figures 4–6 were used. Summary information for unedited data for all molecules in a grid is displayed by clicking the light blue cell at the upper left as shown in A and C. Note that the numbers need to be adjusted to accommodate cells with manually-changed sequencing calls. Left-clicking on the light blue header cell for each row, except row 1, shows information associated with that specific molecule as shown in B.

Image options and data export. (A) Various features for generating and labeling exported images are selected from the shown drop-down menu. In this image, the default image resolution has been reset from 100 to 600 dots per inch (dpi). (B) Window launched by Image options > Select range of sites shown in images. Shown are the default settings, which can be changed, for example, to remove sites not aligned at the beginning or end of aligned sequences. (C) Export data options are in the shown drop-down menu.

MethylViewer analyses MAPit data using degenerate DNMT probe. (A) Publication-quality, scaled *.svg image of bisulfite-converted PHO5 sequences obtained from cells with uninduced (top panel) or estrogen-induced (bottom panel) expression of chromatin probe M.CviPII. Colors are as in other figures, except that CCD and m⁵CCD sites are indicated by filled white and filled purple triangles, respectively. (B) Edited grids used to export images in A. The settings for Base calling cutoff and Word size, were 200 and 15, respectively. (C) Overall methylation frequencies at each CCD site are indicated below the map showing placement of: positioned nucleosomes (ellipses labeled N−1, N−2 and N−3; Almer and Hörz, (82); upstream activating sequences at which Pho4 binds [red-filled circles labeled UASp1 and UASp2 as mapped by Vogel and Hinnen (83)]; the compound Mcm1-Fkh site [cyan-filled rectangle labeled UASm as described in Pondugula et al. (81)]; and TATA box (white-filled square); major TSS (bent arrow). According to the convention used in budding yeast, base pair coordinates are indicated relative to the first nucleotide of the ATG translational initiation start codon.

FASTA file format. The wild-type sequence and proper strand of the region of interest must appear as the first entry. Names of individual aligned sequences can include any number of alpha-numeric characters; however, the program will reduce the font size of labels in exported images to fit within the allotted space. Users must be certain to erase all non-alpha-numeric characters from the file, except > and <.

MethylViewer interactive data grids. (A) Data key. (B–E) Interactive grids showing raw, unedited methylation data processed from *.ab1 files via Analysis > Interactive view … > CG and GC for HCT116 nuclei treated with 0 U M.CviPI (B), HCT116 nuclei + 100 U M.CviPI (C), RKO nuclei + 0 U M.CviPI (D) and RKO nuclei + 100 U M.CviPI (E). The shown grids were generated with the default Base calling cutoff of 20 and maximum alignment Word size of 15. Each column, except for column 1, contains data for a specific potential methylation site in the reference sequence. Each row, except for the header row 1, represents the unedited sequencing calls (C or T) for each of four cloned and sequenced molecules. Placing the cursor over any cell in the grid, except for those in the header column 1, displays the corresponding sequence of the scored site in a small pop-up window as shown in B, D and E. An alternate way to view sites is to select View > Identify methylation sites, which displays the colors of each methylated dC site as shown in C. In addition, clicking on a light blue or color-coded header cell in row 1, except the first cell, displays specific information about a site as shown in D. Lastly, left-clicking on any cell in the grid displays the file number and sequence of that scored site (C or T) as seen in E. Pop-up windows and sub-menus only obscure white cells with unmethylated residues.

Bisulfite conversion status of dC sites not in queried methylation sites. (A) Visualization by View > Show dC conversion map. Placing the cursor over any nucleotide position in the map displays its exact position in base pairs relative to base pair 1 in the reference in a pop-up window. Residues that sequenced as C (G on reverse strand) and lie outside of queried methylation sites are marked by a vertical blue line. Black and gray vertical lines indicate residues within target methylation sites that are methylated and unmethylated, respectively. (B) Clicking on any nucleotide as shown in A displays the local sequencing trace and alignment to which it is linked.

MethylViewer maps any input methylation specificity. The shown window was activated by Analysis > Interactive view … > Methylation sites > Custom to determine frequencies of non-overlapping methylation in the MAPit data of Figure 7. The default CG: 1 and GC: 2 sites were deleted; the settings in each field were input as shown, and site NCG: 2 was entered by clicking Add. Checking the box for GCG eliminates the overlapping site from the analysis. As many as four sites of any sequence containing one or more deoxycytidines, including non-palindromic sites, can be designated.

Publication-quality, scaled images of hMLH1 MAPit data. Edited data from each of the grids in Figure 7 were exported with Image options as shown in Figure 9, via Export data > Save image drawn to scale, and saved using the *.svg option. Nuclei from cell line HCT116 + 0 U M.CviPI (A), HCT116 + 100 U M.CviPI (B), RKO + 0 U M.CviPI (C), and RKO + 100 U M.CviPI (D). The image is identical to that produced by MethylViewer, except that the amplicon map at the top and legend at the bottom were added in Adobe Illustrator and labels at left were edited slightly (increased font size as well as capitalization and italicization as appropriate and to taste). Also inserted were red rectangles and blue ovals, depicting spans of GC sites methylated by exogenously added M.CviPI chromatin probe (according to 2:2 convention, see text) and inferred positions of nucleosomes, respectively. Bent arrows, TSSa and TSSb; translational initiation codon, ATG. The blue bar is to scale and indicates the 147 bp length of DNA within a nucleosome core particle.