The Variants Frequency Table

1.5.1

1.5.1.1

The Variants Frequency Table shows results one Variant per row and one Sample per column (Figure 1‑47). Initially, cells that contain data are white and cells that contain no data are grayed-out (this can happen, for example, if a Sample has no associated Amplicons whose sequence covers the Variant; see below). When an entire row or column is grayed-out, it is moved to the bottom or right of the Table, respectively. This grayed-out scheme is also used by various display features whereby you can filter out the data according to certain criteria, leaving the data of most interest in white cells, in the upper-left area of the Table. (The ‘Compact table’ option from the Variant data display controls can then remove from view all completely grayed-out rows and columns; see section 1.5.2.4)

Figure 1‑47: The Variants Frequency Table

With respect to columns, the Table is divided into two main parts:

•	The first three columns (with blue Header cells) act as “Headers” to the Variant rows:

◦

The ‘Reference’ column gives the names of the Reference Sequences to which the Variants (in the second column) are associated. The rows are initially sorted from top to bottom in alphabetical order of the Reference Sequence names (this applies separately to rows that contain at least one white cell and to the grayed-out rows that appear at the bottom of the Table).

◦	The ‘Variants’ column gives the names of the Variants whose occurrence frequency for each Sample are given in each row. If two or more Variants are associated with any given Reference Sequence, the Variant names are used for second level sorting in the initial display.

◦

The ‘Max’ column displays the maximum frequency observed for each Variant across all the Samples displayed in the Table. This can be a convenient indicator of whether the software detected a given Variant at sufficient frequency in the computed data to warrant further examination in the individual ‘Sample’ columns, to the right.

•

All other columns give the occurrence frequency observed for each Variant for one Sample (identified in the column Header), one column for each Sample. These columns (excluding the first three) are initially sorted from left to right in alphabetical order of the Sample names (this applies separately to columns that contain at least one white cell and to the grayed-out columns that appear at the right end of the Table).

While the first three columns are always visible, you can scroll through the Sample columns using the scroll bar located at the bottom-right of the Table [see Figure 1‑48, which displays the same data as in Figure 1‑47 but in a more expanded display (see section 1.5.2), showing the scroll bar]. As you scroll to the right, the leftmost Sample columns appear to slide behind the first three rows, so you may end up in situations where you display a partial column just after the ‘Max’ column.

Figure 1‑48: The Variants Frequency Table showing the same data as in Figure 1‑47, but in a more expanded form, showing the scrolling feature that applies to the ‘Sample’ columns.

Since Variants are defined in the context of a Reference Sequence, and Samples are associated with Amplicons (which are in turn defined in the context of a Reference Sequence), there may be Samples in your Project that are not valid candidates for a particular Variant scan. This happens if all the Amplicons associated with the Sample are defined relative to a different Reference Sequence than the Variant; or even if they are from the same Reference Sequence, if they do not cover regions of the Reference Sequence where the Variant is defined. It may even be that the Sample was associated with an Amplicon that should have covered the region of variation, but for some reason no actual reads were sequenced that covered the variant (as distinguished from the case where some reads cover the region of variation, but none of them satisfy the constraints given by the Variant’s Pattern). In these cases, the corresponding Sample-Variant cell in the Table will be grayed-out and contain a single dash (‘-’) character.

The Variants tab also has various common features such as a Mouse Tracker and the “Save table snapshot to image file” and “Save Table to Text file” buttons. Note that since each cell can contain up to 6 values (frequency in forward, reverse, and combined reads, and the three associated denominators of those frequencies; see section 1.5.2.2), the spreadsheet may be constructed with multiple columns for each sample to accommodate all these values.

Another useful feature is that if you pause the mouse over any cell of the Table, a screen tip will open providing relevant information about the content of that cell, as follows:

•	Column header cell

◦	Instructions on the right-click options (see section 1.5.1.2)

•	Reference cell

◦	Name of the Reference Sequence

◦	Beginning of its DNA sequence

◦	Reference Sequence Annotation

•	Variant cell

◦	Name of the Variant

◦	“Pattern” of the Variant, in the Variant Definition Syntax (see section 1.3.2.5.2)

◦	Variant Annotation

◦	“Status” of the Variant (see section 1.3.2.5.3)

•	All “Frequency” cells (‘Max’ or ‘Sample’ columns; shown in Figure 1‑46):

•	Frequency (and number of reads) for the combined orientations and for each orientation; and instructions on the right-click options (see section 1.5.1.2)

◦	Name of the Sample

◦	Name of the Variant

◦	“Pattern” of the Variant, in the Variant Definition Syntax (see section 1.3.2.5.2)

◦	“Status” of the Variant (see section 1.3.2.5.3)

1.5.1.2

Organizing Data in the Variants Frequency Table

The Variants Frequency Table contains the summary results of your Amplicon Project provided by the AVA software: the frequency at which each defined Variant was observed in each Sample, in the Read Data Set(s) analyzed, expressed as a percentage of number of reads included in the calculation. As you examine these results, however, it may be convenient to sort the data or to bring the focus on only certain Variants or Samples at a time. This would be especially true in a large Project with many defined Variants and/or many Samples.

To help with this, right-clicking on any column or row “header” cell opens a contextual menu that offers several sorting or filtering options (remember that the cells in the first three columns are all “row headers”; see section 1.5.1.1):

•	If you right-click on the Reference or Variant column header, the contextual menu will include show/ignore options that apply to rows along with reversion options but no column sort options are available (Figure 1-49 A-B).

•	If you right-click on the ‘Max’, column header, the contextual menu will contain hybrid options: the sort options will apply to the given column, but the show and ignore options will apply to rows (Figure 1-49 C).

•

If you right-click on a Sample column header, the options in the contextual menu will apply to that column, with additional options that apply to all the other Sample columns collectively; the sort options sort the values found in the column, effectively reordering the rows of Variant data (Figure 1-49 D).

•

If you right-click on a cell in the column underneath the Reference label, there will be options that apply to all rows collectively and also to that subset of rows that are associated with the specific Reference Sequence: i.e., all those rows that have data for Variants associated with that Reference Sequence (Figure 1-49 E).

•

If you right-click on a cell in the columns underneath the Variant or Max headers, the options in the contextual menu will apply to that row or to all rows of Variants collectively; the sort options sort the values found in the row, effectively reordering the columns of Sample data. The menu also has a Variant Status option that pops up a set of radio buttons for Status selection (Figure 1-49 F).

•	If you right-click on a Sample/Variant intersection cell, the contextual menu contains options to view the Global Alignment from which the frequency data for that cell was derived, and to edit the Variant Status or remove the Variant from the project (Figure 1-49 G).

•

If you right-click on a cell in the columns underneath the Variant or Max headers or if you right-click on a Sample/Variant intersection cell, there is also a “Define Haplotype” option (Figure 1-49 F and G). This option is inactive unless rows for two or more Variants from the same Reference are selected. The “Define Haplotype” option is described in section 1.5.1.4.

A	B	C	D
E	F	G

Figure 1-49: The contextual menus available in the Variants tab (see description above)

The data organization tools offered in these contextual menus include sorting, ignore filters, show filters, and option reversions. These are described below.

1.5.1.2.1

Sort options

•	sort ascending

•	sort descending

These options sort the columns/rows according to the ‘Combined’ Variant frequency in the column/row on which you right-clicked. A blue marker appears in the lower-left corner of the header cell (the ‘Max’ header for rows) in the column/row according to which the sorting was done. The ‘Max’ column header can be sorted and marked just like the Sample columns. You may apply a sort to only one row and only one column at a time, but you can sort the Table according to both one row and one column simultaneously.

Figure 1‑50: The Variants Frequency Table, sorted (descending) according to the Sample2 column. Note the blue marker in the lower-left corner of the Sample2 column header; compare with Figure 1‑47 .

1.5.1.2.2

Ignore filters

•	always ignore column/row

•	ignore all columns/rows

As indicated above (section 1.5.1.1), the software grays-out cells that contain no data and shifts rows and columns that contain only gray cells to the bottom or right ends of the Table. This focuses the cells of interest (white) to the upper-left area of the Table. With the ‘always ignore column/row’ options, you can gray-out any columns/rows (which moves them to the right/bottom of the Table) to help focus on data of current interest. If you want to gray-out most columns/rows, you can use the ‘ignore all’ option to gray them all first, and then apply the ‘show’ filter (see below) to re-focus on only the ones for which you have a current interest. (Cells will also be grayed-out if they fail the Min/Max filter, from the Variant data display controls; see section 1.5.2.3.)

A blue marker appears in the upper-left corner of the header cell (the ‘Max’ header for rows) in the column/row that you chose to ignore, to remind you that this filter was applied manually. You can ignore any number of columns or rows. You can also ignore a column or a row that was used for sorting; in this case, the corresponding header cell will have both blue markers, in its upper- and lower-left corners.

1.5.1.2.3

Show filters

•	always show column/row

•	always show all columns/rows

These filters have the opposite effect of the ‘ignore’ filters (including moving the columns/rows to the upper-left area of the Table). They use the same upper-left blue marker in the header cell as the ‘ignore’ filters, to indicate that the show filter was applied manually to the column/row. They will override the application of an ‘ignore’ filter, as described above, and can be used to force a Sample or Variant into the display even if it has failed the Min/Max filters (see section 1.5.2.3). This can be particularly useful for forcing the inclusion of negative controls in the display (which would typically fail a minimum filter).

1.5.1.2.4

Option reversions

The right-click ‘revert’ options below can be used to selectively undo any of the option choices you have made.

•

‘revert to name sort’ removes the numerical sort on the values of column or row (reverts to the default alphabetical order of the column or row header labels), and causes the lower-left blue marker in the header cell of the column/row that had been used to sort to disappear. You don’t have to click on the specific column/row that was the object of the sorting to get this option; the ‘revert to name sort’ from another column/row would have the same effect.

•	‘auto show column/row’ reverts any show or ignore filters that may have been applied to the column or row on whose header cell you right-clicked. If that column/row was also used for sorting, sorting is not effected.

•	‘auto show all columns/rows’ reverts all show or ignore filters that may have been applied to any column or row (without removing any sorting that may have been applied to the table).

If you want to undo all the sorts and filters you applied (i.e. restore all defaults), use the ‘Reset table’ button.

Button	Name – Description
	The ‘Reset table’ button removes all sorting and ignore/show filters that have been applied to the table and restores the table to its state prior to the filters. Note that this does not affect any of the table formatting options from the Variant data display controls (see section 1.5.2), including the Min/Max filter settings (section 1.5.2.3).

1.5.1.3

Populating the Global Align Tab from the Variants Tab

As described above, right-clicking on a column or row header cell in the Variants Frequency Table opens a contextual menu with the sorting and filter options. By contrast, right-clicking on a cell in the ‘body’ of the Table opens a contextual menu that allows you to populate the Global Align tab with the multi-alignment of the reads belonging to the Sample-Variant combination whose Variant frequency information the cell contains. This is the best way to initiate your exploration of the underlying reads if you are specifically interested in results for a given Variant, as it will selectively include reads from all Amplicons that cover that Variant (and that are associated with that Sample). Such a selection could not be done directly from any of the tree views (see sections 1.3.1 and 1.6.1) because these only allow you to choose a single Sample-Amplicon pair. When you navigate to the Global Align tab from a specific Variant on the Variants Tab, the position of the alignment containing the leftmost position of the Variant Pattern is highlighted to help you visually key in on the variation.

1.5.1.4

Defining a Haplotype from the Variants Tab

Right-clicking on a cell in the columns underneath the Variant or Max headers, or right-clicking on a Sample/Variant intersection cell, reveals a “Define Haplotype” option (Figure 1-49 F and G) in the contextual menu. This option uses a multi-row selection of individual Variants in the Variant Frequencies Table to propose a new Variant that requires that all the selected Variants be found together as a haplotype. This option is grayed-out and inactive unless rows for two or more Variants from the same Reference are selected (a multi-selection of Variant rows from mixed References will inactivate the option even if any of the selected Variants share a Reference). After selecting two or more such Variant rows, right-clicking over any cell in the selection except for a Reference cell provides access to an active “Define Haplotype” option (Figure 1‑51).

Figure 1‑51: The Variants Frequency Table with two individual Variant rows selected. Right-clicking over a non-Reference cell in the selection provides a contextual menu with an active “Define Haplotype” option that can be used to propose a new Variant that requires the selected Variants to be found together as a haplotype.

Choosing the option will bring up the “Define Haplotype” window (Figure 1‑52). This window functions similarly to the “Approve new variant” window used by the “Declare project variant” function of the Global and Consensus Alignment tabs (see section 1.6.3.3, Figure 1-62, and Figure 2-36, below). The main difference between the use of “Define Haplotype” and “Approve new variant” is a matter of context. The “Approve new variant” function is triggered in the context of a multiple alignment where the linkage of Variants in a haplotype can be visually verified. The “Define Haplotype” function is triggered from the main Variant Tab wherein similar Variant frequencies may be suggestive of a haplotype, but which may also simply be coincidental. Accordingly, the “Define Haplotype” window defaults to creating the haplotype with a “Putative” Status.

If the Variants selected as the haplotype constituents have any contradictory elements (e.g., specifying two different SNPs for the same position or specifying both a SNP and a deletion for the same position), then a window similar to that encountered when the user enters erroneous variant patterns (Figure 1‑29) will be displayed to explain the problem. The user then has the ability to edit the resulting pattern and create a semantically correct haplotype definition; more likely, however, one would select the Cancel button since any identified errors would actually disprove the coincidental existence of the Variants as a valid haplotype.

Figure 1‑52: The “Define Haplotype” window with a Pattern built from the Variant selections made in Figure 1‑51 . The Status defaults to Putative but can be edited, and a red warning is given to indicate that a variant with the same pattern already exists.

The Auto-Detected Variant system does not propose haplotypes (beyond the special case of multiple base pair deletions), so haplotypes must be entered into the system by the “Define Haplotype” or “Approve new variant” methods, or they must be defined from scratch in the Variant Definition Table. In large projects it may become difficult to keep track of all the proposed haplotypes that have already been entered into the system, so if an attempt is made to use the “Define Haplotype” or “Approve new variant” functions to propose a haplotype that already exists, a red warning message stating the redundancy will appear in the window (Figure 1‑52).

1.5.1.5

Editing/Removing Variants from the Variants Tab

Right-clicking on a cell in the “body” of the Variants Frequency Table, at a Sample/Variant intersection (as above in section 1.5.1.3 and as shown in Figure 1-49 G), provides a contextual menu that includes options for editing the Variant Status and for removing Variants from the project. The same functions can be carried out using the Variants Definition Table in the Variants sub-tab of the Project Tab (see section 1.3.2.5). The Variants Frequency Table is a convenient place to decide on the Status of a Variant because the incidence frequency across Samples is available for examination. You can use the shift or control keys to help select multiple rows and you can delete them or adjust their Status as a bulk operation. In the case of removing Variants you are provided with a “Yes to All” confirmation window (similar to the case described in section 1.3.2 and Figure 1‑16). A bulk Status edit occurs without any confirmation step.

•	The Status is applied to the Variant and not a Sample. You can’t mark the Status for a Variant to “Accepted” for one Sample and “Putative” for another; the Status is applied globally to the Variant regardless of the Sample(s) in which it is found.

•

If you remove a Variant from the Project it gets removed from every Sample column in the Variants Frequency Table, not just from the Sample corresponding to the cell on which you clicked. If the Variants you remove happen to be Auto-Detected Variants, they could be re-imported the next time you load the Auto-Detected Variants (depending on your filter settings for the load). For this reason, it is usually better to mark the Variants’ Status to “Rejected” rather than remove them from the Project, until you are sure that you will not be loading any more Auto-Detected Variants into the Project.

1.5.1.6

The Mouse Tracker

The Mouse Tracker on the Variants Tab is slightly more complex than on other tabs because the information it displays is highly context-sensitive:

•

If the mouse is over a header cell in the Variants Frequency Table, the Mouse Tracker gives general information about how many different items (Reference Sequences for the ‘Reference’ column; Variants for all other columns) are present in the column, and how many of these comprise data that meet the current Min/Max filter settings.

•	If the mouse is over a specific ‘Reference’ cell, the Mouse Tracker shows the number of Variants and Samples associated with that Reference Sequence, as well as a count of how many of those Samples comprise data that meet the current Min/Max filter settings.

•	If the mouse is over a specific ‘Variant’ cell, the Mouse Tracker shows how many Samples are in the table for that Variant, as well as a count of how many of those Samples comprise data that meet the current Min/Max filter settings.

•

If the mouse is over a specific Sample-Variant cell, the Mouse Tracker shows a set of frequency statistics for that Sample-Variant combination including the frequencies at which that Variant occurred in this Sample among reads in the forward, reverse, and combined orientations; and corresponding denominators (number of reads covering the Variant position) used in these frequency calculations. All these values are shown in the Mouse Tracker even if the setting of the ‘Show values’ option (see section 1.5.2.2) is more restrictive; this way, you have access to all the information even if you choose a more compact Table view.

•

If the mouse is over a specific ‘Max’ cell, finally, the Mouse Tracker behaves as if you were hovering the mouse over the actual Sample-Variant cell that contains those maximum values. Note that ‘Max’ cells in the Table do not display denominators even when the ‘Show denominators’ option is chosen (section 1.5.2.2), but the Mouse Tracker does.