The “Definition Table” Sub-Tabs

1.3.2

The right-hand panel of the Project tab contains seven sub-tabs, one for each type of element that makes up an Amplicon Project (except “Group” element types). The number of elements listed in a Definition Table appears next to the tab name, in parenthesis. The row for a particular element in a Definition Table can be selected by clicking on any of its cells and you can select multiple rows by holding down the ‘shift’ or ‘control’ keys while making your selections. The Tables on these sub-tabs can be used to create or delete elements of the corresponding type for your Project, using most of the same buttons or right-click functions as for the “Tree” sub-tabs of the left panel (see section 1.3.1).

Button	Name – Description
	To add a new element in the Project (except for Read Data Sets and Groups), either click in the appropriate sub-tab to make it the focus of the application and click the “Add” button to the left of the Project Tab; or right-click on an existing element in the appropriate Definition Table and select the “Add” option from the contextual menu that appears. The “Add” action is used to create new elements that can be completely defined using tools of the Project tab. However, certain data, specifically the Read Data Sets, are data which are defined outside the application and must be imported into the Project. For this data there is the “Import” action.
	To import Read Data Sets in the Project, either click in the Read Data sub-tab to make it the focus of the application and click the “Import data” button to the left of the Project Tab; or right-click on an existing element in the Read Data Definition Table and select the “Import” option from the contextual menu that appears. To import en-masse definitions for other project elements for which there is a sub-tab (References, Amplicons, Samples, Variants, MIDs, Multiplexers) the same mechanism may be used (see section 1.3.1 for a more detailed discussion of this feature).
	To remove an existing element from the Project (including a Read Data Set), either select it in its Definition Table and click the “Remove from project” button to the left of the Project Tab; or right-click on the element to be deleted in its Definition Table and select the “Remove” option from the contextual menu that appears. A confirmation window will appear; click “Yes” to delete the element. If you make multiple selections and choose the “Remove” action, the confirmation window will prompt you to remove each one individually, or you can click the “Yes to all” button to confirm the removal of all selected elements at once (Figure 1‑16).
	The “Duplicate item” button is used to create copies of items in the Definition Tables. This is another contextual button that operates on an item that is selected in one of the Definition Table sub-tabs. Clicking on this button while a single item is selected in the Definition Table will add an extra row to the table that is identical to the selected item except that its name will have a suffix of the form “_copy_NUM” (where “NUM” increments from 1 when more than one copy is made of an original item). A copy of a copy adds another copy suffix (e.g., “ItemName_copy_2_copy_1” would result from a duplication of “ItemName_copy_2”). The duplication operation only duplicates data that is explicitly associated with the item in the Table row: it does not duplicate any associations the item might have as implied by the tree structures (such as Sample-Amplicon associations) unless they are specified in the Table (such as the Reference association in the Amplicon and Variant Definition Tables, and the content of Multiplexers). The duplication of Read Data is not currently supported.

Figure 1‑16: The Amplicons Definition Table with a multiple selection applied to rows. The user is prompted to remove the selected rows one at a time, which can be done by clicking “Yes” to each; or you can remove all the selected rows at once using the “Yes to All” button.

Contrary to the case with the tree tabs, the only associations you can create or modify within the element Definition Tables are the ones between:

•	Amplicons or Variants and their Reference Sequence,

•	Read Data Sets and their Read Data Groups, or MIDs and MID Groups, which are done via a drop down menu that appears when you double-click in the corresponding cell, in these Tables

•

The triads Amplicons / Read Data Sets / Samples (indirectly), when MIDs and Multiplexers are used. Specifically, modifying the MID and Sample associations for a given Multiplexer, using the functionality found in the Multiplexer Definition Table, will dynamically update the associated Samples for any Amplicons of a Read Data Set that are associated with the changed Multiplexer.

The “Remove association and remain in project” button or its right-click equivalent are never available when the focus of the application (the last place you clicked) is on a sub-tab of the right-hand panel because not all associations are explicit in the Definition Tables (especially the multiple associations an element may have); indeed, the ability to view the network of element associations in the Project is one of the main benefits of the Project Tree views.

However, associations can also be established by dragging an element from its Definition Table on a right-hand sub-tab, to an appropriate element on a tree view in a left panel sub-tab: hold the mouse button until the name of the “destination” element on the tree view turns green, and then release the mouse button (Figure 1‑17).

↓

Figure 1‑17: Creating an association by dragging an element from its Definition Table on a right-hand sub-tab to an appropriate element on a tree view in a left panel sub-tab. In this case, Sample6 (with its two previously-associated Amplicons EGFR_21_1 and EGFR_21_2) are being associated to Read Data Set DGVS90J04. (See the “Caution” in section 1.3.1.2 for special information about dragging Samples into the Read Data Tree.)

•	The directionality of the dragging capability is always from a Definition Table on the right to an appropriate tree node on the left you cannot establish an association by dragging a tree node object to an object in its Definition Table.

•	The software will not allow you to establish invalid associations, such as linking an Amplicon to a Variant: only elements that are valid destinations for the element you are dragging will turn green and allow the creation of the new association.

•

If you drag one (or more) valid element(s) to the root node (the “Project” folder) in the Samples Tree, or to the root node, a Read Group node or a Read Data node in the Read Data Tree, the element will become associated with all the relevant elements in the nodes below and will be added to all the corresponding branches of that tree (unless this would cause an Amplicon to become associated with more than one Sample or Multiplexer within a Read Data Set branch of the Read Data Tree; in such a case, the existing association remains and only new, non-conflicting associations are created).

◦

This is particularly useful when the experimental design requires the association of one or more Amplicons to a large number of Samples, in one or more Read Data Sets. Such a design may be especially common for experiments using MIDs, where the same Amplicon(s) are to be monitored in multiple MID-tagged libraries (Samples). In this case, Amplicons dragged to an upper node will “trickle down” and become associated with all the relevant and eligible objects below that node. However, the restriction that an Amplicon be associated with only one Sample (or Multiplexer) within a given Read Data Set imposes the following behavior for this function:

▪

The “trickle” will only be accepted if there is at least one Read Data at or below the level of the drag that has at most one Sample or Multiplexer currently associated with it [so it is unambiguous which Sample or Multiplexer the dragged Amplicon(s) should be associated with] and if the Amplicons are not already associated with those Samples or Multiplexers.

▪	If multiple Amplicons are dragged together and some of them are already associated with some of the Samples or Multiplexers under the recipient node, but others aren’t, then the non-associated Amplicons will become associated, and the others will be ignored.

In general, to add or edit the information for an element in its Definition Table, double-click in the corresponding cell in the Table. In some cases, the content of the cell will be highlighted and you can type in the new content; in other cases, the double-click opens a separate window for data entry. Two characteristics, Name and Annotation, are common to all element types; the method to enter or edit them is provided below. The editing methods applicable to the other characteristics of each element type are specified in the sub-sections below.

To edit a Project element’s Name

1.	Double-click in the “Name” cell for the element you want to re-name, in its Definition Table.

2.	Overtype the new name.

3.	Press “Enter” or click elsewhere.

To enter or edit a Project element’s Annotation

1.	Double-click in the “Annotation” cell for the element you want to re-name, in its Definition Table. An “Edit Annotation” window will open (Figure 1‑18).

2.	Type or overtype the information desired.

3.	Click “OK”.

Figure 1‑18: The Edit Annotation window, used to enter or edit the Annotation from any type of Project element

Each Definition Table sub-tab has a header row that labels the content type of each column. The data in the table can be sorted by column content. A black triangle indicator appears to the right of the header label of a column if a sort has been applied to it: an upward-pointing triangle indicates that an ascending sort has been applied to the table data, while a downward-pointing triangle indicates a descending sort. Keep in mind that not all sorts are purely alphabetical or numerical; Project element Names, in particular, are broken down into numerical and non-numerical sections and the sections are sorted either numerically or alphabetically as appropriate.

Clicking on a column header that does not currently have a sort applied to it will cause an ascending sort of the table based on the data in that column. Clicking on a column header that already has an existing sort applied to it will toggle the sort type (ascending or descending). Only one column may have an active sort applied to it at a time, but the sort operations are stable and maintain the prior table order in cases where ties are encountered during the sort operation. This allows you to apply more than one sort at a time. For instance, if you wanted to sort the Variants Definition Table (Figure 1‑19A) by status with the entries subsorted by variant name, you would first click on the ‘Name’ column header to sort by variant name (Figure 1‑19B) and then you would click on the ‘Status’ column to sort the entries by status (Figure 1‑19C).

Figure 1‑19: (A) A Variants Definition Table with unsorted entries. (B) A Variants Definition Table with entries sorted by name. Clicking on the ‘Name’ header performs an ascending sort of the rows as indicated by the upward-pointing black triangle. (C) A Variants Definition Table with entries sorted by status. Clicking on the ‘Status’ header performs an ascending sort of the rows as indicated by the upward-pointing black triangle. Since an ascending name-sort was performed first (panel B), the rows within each ‘Status’ category have an ascending name-sort.

If you add a new entity to a Definition Table and there is a sort applied to a column for which the new row has a default value (such as the ‘Name’ field), the new row will be inserted into the table in its proper place according to the default value and the sort order. If you edit a cell in a sorted column and you change the value to something that would cause the row to change its position according to the sort order, however, sorting is automatically turned off for that column. This behavior prevents the row from moving out from under you as you enter an edit in a cell in a sorted column.

When the sorting is turned off, none of the header labels for the table will have the black triangle indicator. You can re-enable sorting by clicking on any column header you want to sort by. If the first column header you click on for sorting is the same one that was used for sorting just prior to sorting deactivation, you will restore the same sort order (ascending or descending) that was last used for the column. Clicking on any other column header will result in the default ascending sort for that column.

1.3.2.1

The References Definition Table

The References Definition Table lists all the Reference Sequences defined in the Project, with the following three characteristics (Table columns; see Figure 1‑20):

•

Name

•	Annotation (free user-entered text)

•

Sequence

Figure 1‑20: The References Definition Table sub-tab of the Project Tab’s right-hand panel

For the procedures to add or remove Reference Sequences in a Project, see section 1.3.2 (or 1.3.1, to accomplish this in a “Project Tree” view). For the procedures to enter/edit the Name or Annotation information for a Reference Sequence, see section 1.3.2. The sub-section below provide the procedure to enter/edit the other characteristic of Reference Sequences, the DNA sequence itself.

1.3.2.1.1

To Enter or Edit the DNA Sequence of a Reference Sequence

1.	Double-click in the “Sequence” cell of the Reference Sequence you are defining, in its Definition Table. An “Edit Sequence” window will open (Figure 1‑21).

2.	Paste or type the sequence (only A, T, G, C, or N characters; see Caution, below).

3.	Click “OK”.

Characters restriction: Be aware that only “nucleotide” characters (A, T, G, C, or N) are accepted when you enter a Reference Sequence into the AVA software (by typing or pasting). For convenience, when pasting sequences, characters that are not nucleotide characters and are also not IUPAC ambiguity characters (such as R for purine, Y for pyrimidine, etc.) are removed from the pasted entry. This is useful when pasting sequences from sources that may include non-sequence information (such as white space or numerical position information in the margin of each line). During such pastes, any IUPAC ambiguity characters are converted to “N” characters, as the other ambiguity characters are not supported by the software (typing individual “ambiguous” characters, however, does not result in their conversion to “N”; these are simply ignored and the text “Only ATGC and N” at the top of the Edit Sequence window turns bold and red to alert you that an invalid character was used). The restriction that no ambiguity characters other than N be present in a sequence is a requirement of many alignment algorithms and is not unique to the 454 Sequencing System software.

Figure 1‑21: The Edit Sequence window, used to enter or edit the DNA sequence of a Reference Sequence element

1.3.2.2

The Amplicons Definition Table

The Amplicons Definition Table lists all the Amplicons defined in the Project, with the following seven characteristics (Table columns; see Figure 1‑22):

•

Name

•	Reference Sequence (to which the Amplicon is associated)

•	Annotation (free user-entered text)

•	Primer 1 (5’-->3’ sequence of the forward primer of the Amplicon)

•	Primer 2 (5’-->3’ sequence of the reverse primer of the Amplicon)

•	Start (first nucleotide of the Target, on the Reference Sequence)

•	End (last nucleotide of the Target, on the Reference Sequence

Figure 1‑22: The Amplicons Definition Table sub-tab of the Project Tab’s right-hand panel

For the procedures to add or remove Amplicons in a Project, see section 1.3.2 (or 1.3.1, to accomplish this in a “Project Tree” view, and concurrently create associations). For the procedures to enter/edit the Name or Annotation information for an Amplicon, see section 1.3.2. The sub-sections below provide the procedures to enter/edit the other characteristics of Amplicons.

1.3.2.2.1

To Enter or Edit the Reference Sequence to which an Amplicon is associated

Ensure that the column labeled “Reference” in the table is wide enough to allow you to distinguish among the Reference Sequences from which you want to select. The column may be widened by clicking on the separator line, in the table header, between the Reference and Annotation columns, and dragging the separator to the right.

2.	Double-click in the “Reference Sequence” cell for the Amplicon you are defining, in its Definition Table. A drop down menu will expand, showing all the Reference Sequences currently listed in the Project.

3.	Select the proper Reference Sequence from the drop down menu. The new association will automatically appear on the References Tree, on the left panel.

If the Reference Sequence does not yet contain a DNA sequence (see section 1.3.2.1.1), you will still be able to associate Amplicons to it, but you will not be able to fully define them. In particular, you will not be able to specify the Target Start and End for the Amplicons (see section 1.3.2.2.3, below) because these are set using the position numbering from the Reference Sequence.

1.3.2.2.2

To Enter or Edit the Primer Sequences for the Amplicon

As mentioned earlier (section 1.1.1.3), Primer 1 and Primer 2 correspond to the “sequence-specific” part of the two Fusion Primers used to construct the Amplicon library, excluding the 19 bp “Primer A” and “Primer B” parts of the Fusion Primers.

•	Both Primer 1 and Primer 2 should be entered as their true 5’-->3’ sequence. To find the End of the Target (section 1.3.2.2.3, below), the software automatically determines the reverse-complement of Primer 2 (Primer 2’) and aligns this to the Reference Sequence.

•	The AVA software does not require any knowledge of A vs. B beads from the emPCR Amplification kits; reads that align in the same orientation as the given Reference Sequence are considered ‘forward’ reads, and those that must be reverse complemented to align are considered ‘reverse’ reads.

1.	Double-click in the “Primer 1” (or “Primer 2”) cell for the Amplicon you are defining, in its Definition Table. An “Edit Primer 1” (or “Edit Primer 2”) window will open (Figure 1‑23).

2.	Paste or type the sequence (only A, T, G, C, or N characters; see Caution, below).

3.	Click “OK”.

Characters restriction: Be aware that only “nucleotide” characters (A, T, G, C, or N) are accepted when you enter a Primer Sequence into the AVA software (by typing or pasting). For convenience, when pasting sequences, characters that are not nucleotide characters and are also not IUPAC ambiguity characters (such as R for purine, Y for pyrimidine, etc.) are removed from the pasted entry. This is useful when pasting sequences from sources that may include non-sequence information (such as white space or numerical position information in the margin of each line). During such pastes, any IUPAC ambiguity characters are converted to “N” characters, as the other ambiguity characters are not supported by the software (typing individual “ambiguous” characters, however, does not result in their conversion to “N”; these are simply ignored and the text “Only ATGC and N” at the top of the Edit Sequence window turns bold and red to alert you that an invalid character was used). The restriction that no ambiguity characters other than N be present in a sequence is a requirement of many alignment algorithms and is not unique to the 454 Sequencing System software.

Although the AVA software allows “N” characters in the primer sequences, the primer trimming and demultiplexing computational steps perform better if only A, T, G, or C characters are used. If your primer design involves “wobble" positions in which more than one base may appear, it is preferable to define the primer sequence with one of the alternative bases rather than an N at those positions. For record keeping purposes, you may document the choice of data entry in the corresponding Amplicon’s Annotation field.

Figure 1‑23: The Edit Primer 1 window, used to enter or edit the DNA sequence of Primer 1 for an Amplicon element

1.3.2.2.3

To Enter or Edit the Target Start and End Positions

There is no requirement that the Reference Sequence contain either of the two Primer sequences, although they typically will be contained therein. The Target, however, is required to be present in the Reference Sequence, and the Target Start and End positions indicate the first and last bases of the Target, inclusive, relative to the Reference Sequence.

To simplify the setting of the Start and End values, and reduce the risk of data entry errors, the AVA software searches for the Primers within the Reference Sequence, on the assumption that they will most likely be present and, if found, uses the primer positions to establish default values for the Target Start and End (see below). As such, prior to entering or editing the Target Start and End positions, the sequence of the Reference Sequence with which the Amplicon is associated and the two Amplicon Primers must already be defined (sections 1.3.2.1.1 and 1.3.2.2.2).

1.	Double-click in either the “Start” or the “End” cell for the Amplicon you are defining, in its Definition Table. Clicking in either the “Start” or the “End” cell opens the same “Edit Start/End” window (Figure 1‑24). This window includes:

a.	a brief set of instructions

b.	a pair of data entry boxes for the Start and End nucleotides

the Reference Sequence to which the Amplicon is associated, with a color-coded overlay for the Primer sequences (matched and mismatched), the Target sequence (the part of the Reference Sequence between the two Primers), and the “unused” sequence (the part of the Reference Sequence outside the two Primers)

d.	the sequence of the two Primers, plus the reverse-complement of Primer 2 (Primer 2’)

Figure 1‑24: The Edit Start/End window, used to define the start and end of the Target for an Amplicon element (the part excluding the Primers), by locating the Primers 1 and 2 on the Reference Sequence with which the Amplicon is associated

2.	There are 3 ways to set (or reset) the Start and End nucleotides of the Target:

If the Target’s Start and End have not been specified for this Amplicon before (i.e. the Start and End cells were empty when you double-clicked them), the software automatically searches for the Primers (Primer 1 and Primer 2’) in the Reference Sequence; if it finds them (exact matches only), the software marks the Primers in yellow and the Target sequence (between the two Primers) in blue, and specifies default values for the Target’s Start and End positions in the boxes at the top of the window. The user should verify that the default positions are correct since, in some rare circumstances, there may be multiple Primer1-Primer2’ pairs of matches within the same Reference Sequence and the software simply gives the first such pair it finds. This Primer search function can also be elicited by typing a “0” (or a negative number) in either the Start or the End entry box. It is possible that exact matches for the Primers are not found in the Reference Sequence, as either or both Primers may actually not be represented by the Reference Sequence or, due to design considerations (or primer synthesis or sequencing errors), the Primers may slightly differ from the Reference Sequence so that they have a close, but inexact match. Whatever the reason, if no exact match can be found for Primer1, the AVA software will default the Target Start to the first base of the Reference Sequence; if no exact match can be found for Primer2’, the default for Target End will be the last base of the Reference Sequence. If this happens, verify that you have correctly defined the Primer and the Reference Sequence to which this Amplicon is associated; if the sequences are correct, but the default values supplied are incorrect, use one of the following methods to specify the Target Start and End positions.

If you know the exact positions of the Target’s Start and End relative to the Reference Sequence, you can type them in the entry boxes at the top of the window. The AVA software will automatically update the color-coded display to indicate how well Primer1 and Primer2’ match to the Reference Sequence in the regions abutting the supplied Start and End positions. If you know that the Reference Sequence doesn’t actually contain the Primers themselves, then you can safely ignore this feedback. However, if the Reference Sequence is supposed to contain the primers and there are one or more bases of mismatch indicated by pink highlighting in the display, you should check that you entered the Start and End positions correctly, that you entered the correct Primers (and that both are in the 5’-->3’ orientation), that you have the correct Amplicon-Reference sequence association, and that the Reference Sequence itself has the correct sequence.

You can also use the mouse drag method: the software interprets click-and-dragging the mouse in the sequence as an attempt at selecting the amplified range (the Target), so it aligns the sequence beyond the drag point with the Primers (Primer 2’ if dragging to the right and Primer 1 if dragging to the left). Matching nucleotides are overlaid in yellow and non-matching nucleotides, in pink. This method may be especially useful if the Primers you used to generate the Amplicon library did not exactly match the Reference Sequence you are using for analysis. To use this method, do the following:

Click near the beginning of the Reference Sequence and drag the mouse to the right. The stretch of sequence (the length of Primer 2’) beyond the drag point will be displayed in color to indicate matches with the Primer. Stop and release the mouse button when the whole stretch is yellow, indicating a perfect match with Primer 2’ (or allow for mismatches, shown in pink, if appropriate). The software enters the start and stop points of the dragging action in the Start and End entry boxes, respectively; this sets the Target’s End nucleotide.

ii.

Click again on the last nucleotide of the Target, as just defined, and drag the mouse to the left. Stop and release the mouse as before, when you have located Primer 1. Again, the software enters the start and stop points of the dragging action, but in the End and Start entry boxes, respectively; this sets both the Target’s Start and End nucleotides.

3.	Click “OK”.

1.3.2.3

The Read Data Definition Table

The Read Data Definition Table lists all the Read Data Sets defined in the Project, with the following four characteristics (Table columns; see Figure 1‑25):

•

Name

•	Annotation (free user-entered text)

•	Group (the Read Group to which the Read Data Set belongs)

•	Active (if checked, the Read Data Set will be included in the next computation of the Project)

Figure 1‑25: The Read Data Definition Table sub-tab of the Project Tab’s right-hand panel

For the procedures to add or remove Read Data Sets in a Project, see section 1.3.2 (or 1.3.1, to accomplish this in a “Project Tree” view, and concurrently create associations). For the procedures to enter/edit the Name or Annotation information for a Read Data Set, see section 1.3.2. The sub-sections below provide the procedures to enter/edit the other characteristics of Read Data Sets.

The Read Data files generated by the standard data processing pipelines are named uniquely, and within the files, the read accnos (which are based on the file name) are also unique. Read Data files with duplicate names are not allowed in AVA. So, if unmanipulated Read Data files are imported into an AVA project, all reads and their accnos in the project will be unique.

However, manipulation of Read Data files by either renaming them or using SFF Tools to merge or filter them, can result in situations where reads are present in a project in duplicate (and depending on the project, the duplicates may appear in the same alignment). If the Read Data files are manipulated, care should be taken to ensure that reads are not unintentionally duplicated within a project.

1.3.2.3.1

To Edit the Read Group of a Read Data Set

If you want to transfer a Read Data Set to another pre-existing Read Data Group, double-click the drop down menu in the Group cell for the Read Data Set and select the Read Group you want from the available choices. You can also reassign a Read Data Set to a different Read Group by dragging the Read Data Set to a Read Group node of the Read Data Tree (a multiple-selection of Read Data Sets will assign them all to the Read Group on which you drop them). While you cannot change the name of a Read Group from within the Read Data Definition Table, you can do so in the Read Data Tree (as with any other rename operation in the tree, click once on the Group name, pause and click a second time to activate the name editor). Note that all Read Groups are distinct entities, and although you can rename an existing Read Group to match the name of another pre-existing Read Group, this will not cause the Read Data Sets to be merged into the same group.

Note also that you cannot import the same Read Data Set more than once in a Project, even if you intend to assign them to different Read Groups. If you attempt to do so, an error message will appear on screen.

1.3.2.3.2

To Edit the “Active” status of a Read Data Set

You may elect to include or exclude from future computations, any Read Data Set defined in the Project. Excluding a Read Data Set from future computations does not remove it from the Project. If you discover that a Read Data Set is unsuitable for the Project in some manner, it can be useful to keep it in the project but mark it as inactive (and possibly update its Annotation) to serve as a reminder not to reintroduce the data at some future point.

1.	To include a Read Data Set in future computations of the Project, check its box in the “Active” column of the Read Data Definition Table.

2.	To exclude a Read Data Set in future computations of the Project, uncheck its box in the “Active” column of the Read Data Definition Table.

1.3.2.4

The Samples Definition Table

The Samples Definition Table lists all the Samples defined in the Project, with only the following two characteristics (Table columns; see Figure 1‑26):

•

Name

•	Annotation (free user-entered text)

Figure 1‑26: The Samples Definition Table sub-tab of the Project Tab’s right-hand panel

For the procedures to add or remove Samples in a Project, see section 1.3.2 (or 1.3.1, to accomplish this in a “Project Tree” view, and concurrently create associations). For the procedures to enter/edit the Name or Annotation information for a Sample, see section 1.3.2.

1.3.2.5

The Variants Definition Table

The Variants Definition Table lists all the Variants defined in the Project, with the following five characteristics (Table columns; see Figure 1‑27):

•

Name

•	Reference Sequence (with which the Variant is associated)

•	Annotation (free user-entered text)

•	Pattern (definition of the nature of the Variant)

•	Status (workflow category)

Figure 1‑27: The Variants Definition Table sub-tab of the Project Tab’s right-hand panel

For the procedures to add or remove Variants in a Project, see section 1.3.2 (or 1.3.1, to accomplish this in a “Project Tree” view, and concurrently create associations). For the procedures to enter/edit the Name or Annotation information for a Variant, see section 1.3.2. The sub-sections below provide the procedures to enter/edit the other characteristics of Variants.

1.3.2.5.1

To Enter or Edit the Reference Sequence to which a Variant is associated

2.	Double-click in the “Reference Sequence” cell for the Variant you are defining, in its Definition Table. A drop down menu will expand, showing all the Reference Sequences currently listed in the Project.

3.	Select the proper Reference Sequence from the drop down menu. The new association will automatically appear on the References Tree, on the left panel.

If the Reference Sequence does not yet contain a DNA sequence (see section 1.3.2.1.1), you will still be able to associate Variants to it, but you will not be able to fully define them. In particular, you will not be able to specify the Pattern for the Variant (see section 1.3.2.5.2, below) because this is set using the position numbering from the Reference Sequence.

1.3.2.5.2

To Enter or Edit the Pattern of a (Known) Variant

If you already know one or more Variants (e.g. from the scientific literature or from previous experiments), you can define them in the Project and have the AVA software report on the frequency at which they occur in the Read Data Sets included in the Project. Note that novel Variants observed in the reads of the Project itself can also be defined as described below, but the best way to specify novel Variants is to examine the multiple alignments of the putative Variants found by the AVA software during computation and to “Accept” them if you determine that they are legitimate (see section 1.3.2.5.3, below); also, you can “declare” novel Variants not identified by the software after you identify and evaluate them in the Global Align or Consensus Align tabs (see sections 1.6 and 1.7).

The AVA software uses 4 types of constraints to define Variants, and writes them following a strict Variant Definition Syntax, summarized in Table 1‑1. A Variant can be specified by one or more constraints which, collectively comprise the “Pattern” that defines the Variant.

Constraint Type	Syntax	Description
Must match	m(p) or m(p1-p2)	A read satisfies this constraint when the nucleotide(s) at position “p” or in the range “p1-p2” (inclusive) of the Reference Sequence are identical to those of the Reference Sequence.
Substitute base	s(p,n)	A read satisfies this constraint when the nucleotide at position “p” is “n” (where “n” is different from the nucleotide at that position in the Reference Sequence).
Insert bases	i(p.5, n…)	A read satisfies this constraint when the (one or more) nucleotide(s) “n…” are present between positions “p” and “p+1” of the Reference Sequence. Note that a deletion may not also exist at positions “p” or “p+1”, as this combination of insertion and deletion would rather define a substitution.
Delete bases	d(p) or d(p1-p2)	A read satisfies this constraint when the nucleotide(s) at position “p” or in the range “p1-p2” (inclusive) of the Reference Sequence are absent. Note that directly neighboring insertions may not also exist, as this combination would rather define a substitution.

Table 1‑1: The language of variations in the AVA software: the Variant Definition Syntax. A Variant may comprise multiple constraints (though any given nucleotide may only have a single constraint) enabling the encoding of haplotypes or other complicated variation Patterns. In these more complicated cases, the Variant is encoded by specifying multiple, concatenated constraints, optionally separated by whitespace.

To enter or edit the Pattern defining a Variant in the Variants sub-tab of the Project Tab, do the following:

1.	Double-click in the “Pattern” cell for the Variant you are defining, in its Definition Table. An “Edit Pattern” window will open (Figure 1‑28). This window includes:

a.	a “Pattern” data entry box to define and view the nature of the Variant, using the Variant Definition Syntax (see above)

b.	a box containing a DNA sequence based on the Reference Sequence to which the Variant is associated, with a color-coded overlay for the graphical definition and visualization of the Variant

Figure 1‑28: The Edit Pattern window, used to specify the difference(s) compared to the Reference Sequence that define the Variant

2.	There are 2 ways to set (or reset) the definition of a Variant, using this window:

You can enter it directly in the Pattern box at the top of the window, using the Variant Definition Syntax (see above). The software automatically adds variations entered in the Pattern box to the graphic sequence box, below it. If the syntax used is incorrect, the AVA software parses the entry and suggests a correction or provides a tip, in the area below the sequence box

You can enter it graphically by selecting nucleotide(s) in the sequence field and assigning the appropriate type of constraint using the buttons to the left of the sequence box. The software automatically adds constraints entered graphically, in the Pattern box. There are 4 buttons, matching the four types of constraints, and a fifth button for clearing a previously specified constraint, that can be used to graphically define a Variant. The 5 buttons, and their use, are as follows:

i.	Must match (shown in yellow overlay)

(a)	Select one nucleotide (click) or a nucleotide range (click-and-drag) in the sequence.

(b)	Click the Must match button.

ii.	Substitute base (shown in pink overlay)

(a)	Select one nucleotide (click) in the sequence.

(b)	Click the Substitute base button; the “One base” window will open (not shown).

(c)	Type the substituting nucleotide.

(d)	Click OK; the sequence changes to that of the Variant.

iii.	Insert bases (shown in blue overlay)

(a)	Select the one nucleotide (click) in the sequence before which the insertion is located.

(b)	Click the Insert bases button; the “Enter insert sequence” window will open (not shown).

(c)	Type the nucleotide(s) to be inserted (only A, T, G, or C characters).

(d)	Click OK; the insertion appears in the sequence. The position of the inserted nucleotides use decimals so that the original Reference Sequence positions are maintained (e.g. position 66.5 means that the insertion is between the nucleotides at positions 66 and 67 of the Reference Sequence).

iv.	Delete bases (shown in gray overlay)

(a)	Select one nucleotide (click) or a nucleotide range (click-and-drag) in the sequence.

(b)	Click the Delete bases button; the nucleotides in the sequence are replaced with dashes.

v.	No constraint (shown with white background; the Reference Sequence)

(a)	Select one nucleotide (click) or a nucleotide range (click-and-drag) in the sequence that already have one of the above changes assigned

(b)	Click the No constraint button; the nucleotide(s) in the sequence revert to the Reference Sequence. (This function is useful if you incorrectly entered a constraint in the definition of a Variant.)

Click OK.

If an erroneous pattern is directly entered into the Pattern field of the “Edit Pattern” window, the AVA software will remove portions of the pattern until what remains is valid both syntactically and semantically, as seen in Figure 1‑29. In this case, one or more parsing error messages will appear describing the nature of the problem, in the context of the full, erroneous pattern entered.

Figure 1‑29: The Edit Pattern window, with an error in the pattern specification. The user attempted to specify both a substitution at positions 97 and 126 as well as a deletion spanning positions 95-98. Since position 97 cannot both be deleted and involved in a substitution, the software automatically removed the deletion from the Pattern specification at the top of the window, but shows the full erroneous pattern with an appropriate error message in an error window. The compatible substitution at position 126 is left as part of the pattern.

1.3.2.5.3

To Edit the Status of a Variant

Variants exist within a Project, with one of three possible Status values: “Accepted”, “Putative”, or “Rejected”. Variants defined manually by the user (see section 1.3.2.5.2) receive the “Accepted” status by default. By contrast, variations between the Read Data Sets and the References that are identified by the AVA software (during computation; see section 1.4), are initially proposed as “Putative” Variants. After you have examined the data underlying a Variant and determined whether you believe it to be legitimate or not, you can change its assigned Status as described below.

Note, however, that the main use of the Variant Status feature is as part of a “Discovery Workflow”, on the Variants tab. The purpose of this process is precisely to determine whether the Variants included in the Project appear to be legitimate or not and to mark them as such (“Accepted” or “Rejected”); The Variants tab is also where the control is located to “load” the putative Variants discovered by the software, into the Project. For these reasons, Variant Status assignment might more often be done on the Variants tab than on the Variants sub-tab of the Project tab, as described here. See section 1.5 for a description of the Variants tab and for more details on this Discovery Workflow.

To edit the Status of a Variant in the Variants sub-tab of the Project Tab, do the following:

1.	Ensure that the “Status” column in the Variants Definition Table is wide enough to allow you to distinguish among the Status choices. The column may be widened by clicking on the separator line, in the table header, between the Pattern and Status columns, and dragging the separator to the left.

2.	Double-click in the “Status” cell for the Variant you are defining. A drop down menu will expand, showing the available Status options.

3.	Select the proper Status from the drop down menu.

•	If you open a Project from a prior version of the AVA software which did not have the Status field, the status value for any pre-existing Variants in the Project will be set to “Accepted”.

•

It is often more useful to use the “Rejected” status than to actually remove a Variant from the Project. When you mark a Variant as rejected, you prevent it from being rediscovered by the auto-variant detection process, during further computations of the Project. Variants can safely be removed after you are done adding new data to the Project.

1.3.2.6

The MIDs Definition Table

The MIDs Definition Table lists all the MIDs defined in the Project, with the following four characteristics (Table columns; see Figure 1‑30):

•

Name

•	Annotation (free user-entered text)

•

Sequence

•	Group (the MID Group to which the MID belongs)

Figure 1‑30: The MIDs Definition Table sub-tab of the Project Tab’s right-hand panel

MIDs may be created (and assigned to MID Groups) in the MIDs Definition Table even before the sequence of the MID has been filled in by the user. Such MIDs, without defined sequences, may even be used in the definitions of the Samples encoded by Multiplexers (section 1.3.2.7.1). This flexibility allows users to define of the logical structure of an experiment in advance of knowing the specifics of the MID sequences themselves.

For the procedures to add or remove MIDs in a Project, see section 1.3.2 (or 1.3.1, to accomplish this in a “Project Tree” view). For the procedures to enter/edit the Name or Annotation information for an MID, see section 1.3.2. The sub-sections below provide the procedure to enter/edit the other characteristics of MIDs.

Criteria for good MID Sets; standard and custom MID Groups:

An MID Group (“454Standard”) containing 14 ten-base MIDs is provided and recommended for use when designing multiplexing libraries for the AVA software. This group includes the set of 12 tags known as MID1 through MID12 that are available in kit form to make Shotgun (sstDNA) MID libraries, and includes two additional MIDs not included in these kits (MID13 and MID14). The MIDs of the 454Standard MID Group have been especially selected for the following qualities:

•

Their sequences are as divergent as possible and include only single nucleotide homopolymers, minimizing the risk that eventual sequencing errors would make any of these MIDs look like one of the other MIDs of the Group (which would cause the assignment of the read to a wrong Sample). Indeed, no fewer than six changes (insertions, deletions, and/or substitutions) separate any two MIDs of the 454Standard MID Group.)

•

All MIDs of the Group are of the same length (10-mers, in this case). The AVA software requires that all the MIDs used for a given end of the Amplicons (Primer A/Primer 1 or Primer B/Primer 2) have the same length, so they are on equal footing for error distance calculation purposes. It is permitted, however, to use a different MID length on each side of the Amplicons being handled by a Multiplexer. For instance, custom 5 bp MIDs might be used on the Primer 1 side and 10 bp MIDs might be used on the Primer 2 side.

•	None start with a “G” nucleotide, the last base of the sequencing key, to ensure clear reading of the MID tag.

•	All MIDs are read within 4 or 5 nucleotide flow cycles, to minimize usage of Run flows to read the MID tags and leave as much as possible for the sample sequence.

The software does not constrain the user to only these 14 MIDs, however: any other set of sequences can be designed for use as multiplexing tags, incorporated between the sequencing key and the template-specific primer of the Adaptors used to prepare the Amplicon libraries, and defined as MIDs (with optional custom MID Groups) in the Amplicon Project. This flexibility can be useful, for example, if the user prefers to use shorter MIDs; or if Amplicon libraries already exist that have intrinsic sequences that can be used for demultiplexing; or if the experiment requires the multiplexing of more Samples than can be differentiated with the 14 MIDs of the 454Standard MID Group. If you design your own MIDs, it is recommended that you keep in mind the 4 criteria listed above, for best results.

Amplicon library chemistry requires the inclusion of the MID sequence as part of the Fusion Primers used in library preparation. These must be obtained separately by the user because they contain sequences that are specific to each Amplicon, so MID Adaptors used for Amplicon libraries are not available as kit reagents. A naming convention has been adopted to distinguish between Shotgun (sstDNA) and Amplicon library MIDs; fully uppercase MID names are Shotgun (sstDNA) library MIDs, and initial uppercase Mids are MIDs used in Amplicon libraries. Thus, MID1 and Mid1 refer to the same sequence content for an MID, but are each used in the context of their respective library types; MID1 is an Adaptor available in a kit, while Mid1 is not.

Contrary to the situation with the GS De Novo Assembler and the GS Reference Mapper applications, the number of acceptable “reading errors” in the MIDs is not set by the user in the AVA software. Rather, the software dynamically calculates how many errors can be accepted by analyzing the set of MIDs used and determining how close they are to each other in terms of the minimum number of insertions, deletions, or substitutions that would be required to transform one MID into another.

Include all MIDs used in the experiment: The analysis of “MID closeness" for MID error correction described in the Note above is based on the MIDs specified in the Multiplexer definitions. For the purpose of this analysis it is important to include all MIDs that were actually used in the sequencing phase of the experiment, even if certain of these MIDs correspond to Samples that are not of interest in the particular AVA project. If any used MIDs were not specified in the project, the AVA software could overestimate the amount of allowable error correction as it tries to match reads to the MIDs it knows, which could result in MID “overcorrection” and the mis-assignment of reads to the known MIDs.

Note that one may include MIDs in the Multiplexer definitions without associating them with any Samples. This allows MIDs to be specified for the “MID closeness” analysis without requiring any reads with those MIDs to be explicitly included in the AVA output. This is useful both for MIDs that are known to be in the experiment but whose reads are not desired in the output, as well as for MIDs that are not intended to be in the experiment but that may be present e.g. as a result of cross-contamination from other experiments that used the same MID / Amplicon primer combinations.

1.3.2.6.1

To Enter or Edit the DNA Sequence of an MID

1.	Double-click in the “Sequence” cell of the MID you are defining, in its Definition Table. An “Edit Sequence” window will open (Figure 1‑31).

2.	Paste or type the sequence (only A, T, G, or C characters; see Caution, below).

3.	Click “OK”.

Characters restriction: Be aware that only “nucleotide” characters (A, T, G, C) are accepted when you enter an MID Sequence into the AVA software (by typing or pasting). For convenience, when pasting sequences, characters that are not nucleotide characters and are also not IUPAC ambiguity characters (such as R for purine, Y for pyrimidine, etc.) are removed from the pasted entry. This is useful when pasting sequences from sources that may include non-sequence information (such as white space or numerical position information in the margin of each line). If any IUPAC ambiguity characters are included, the paste will be cancelled entirely, and an error message will be displayed explaining the problem. If you directly type individual “ambiguous” characters, however, or any character other than A, T, G, or C, these characters are simply ignored.

The restriction that no ambiguity characters be present in an MID sequence is crucial because MIDs are intended to designate specific Samples. If you have a degenerate MID design in which multiple MID sequences specify the same Sample, enter all the specific MID sequences into the system and use the Multiplexer Sample editor to specify all the MIDs that encode each Sample (see section 1.3.2.7.3)

Figure 1‑31: The Edit Sequence window, used to enter or edit the DNA sequence of an MID Sequence element

1.3.2.6.2

To Edit the MID Group of an MID

To transfer an MID to another pre-existing MID Group, double-click the drop down menu in the Group cell for the MID and select the MID Group you want from the available choices. You can also reassign an MID to a different MID Group by dragging the MID to an MID Group node of the MIDs Tree (a multiple selection of MIDs will assign them all to the MID Group on which you drop them). While you cannot change the name of an MID Group from within the MIDs Definition Table, you can do so in the MIDs Tree (as with any other rename operation in the tree, click once on the Group name, pause and click a second time to activate the name editor). Note that all MID Groups are distinct entities, and although you can rename an existing MID Group to match the name of another pre-existing MID Group, this will not cause the MIDs to be merged into the same group.

A valid MID Group should contain MIDs with sequences of the same length, and each MID sequence should be distinct. When assigning MIDs with defined sequences to an MID Group, the software will prevent you from making an inconsistent assignment, such as adding an MID with an a defined sequence to an MID Group that already has at least one defined MID sequence of a different length: if dragging to the MIDs Tree, the MID Group node will not activate to allow you to release the dragged MID; if using the drop down menu from the MIDs Definition Table, the drop down menu will only display valid MID Group choices.

Although the AVA application is designed to minimize the possibility of creating inconsistent MID Groups, to allow flexibility in editing, it allows the addition of MIDs with undefined sequences to MID Groups, and the editing of MID sequences even after they have already been assigned to an MID Group. During such editing you are allowed to bring an MID Group into a temporarily inconsistent state, with the assumption that you will eventually fix it prior to computation. If you use inconsistent MIDs when defining Multiplexers, the Multiplexer setup dialogs will provide you with error messages (section 1.3.2.7.2), and you will also be provided with error warnings prior to computation (section 1.4). Ignoring the warnings will prevent the portions of the computation that depend on the faulty Multiplexers from executing.

1.3.2.7

The Multiplexers Definition Table

The Multiplexers Definition Table lists all the Multiplexers defined in the Project, with the following six characteristics (Table columns; see Figure 1‑32):

•

Name

•	Annotation (free user-entered text)

•

Encoding

•	Primer 1 MIDs

•	Primer 2 MIDs

•

Samples

Figure 1‑32: The Multiplexers Definition Table sub-tab of the Project Tab’s right-hand panel

For the procedures to add or remove Multiplexers in a Project, see section 1.3.2 (or 1.3.1, to accomplish this in a “Project Tree” view). For the procedures to enter/edit the Name or Annotation information for a Multiplexer, see section 1.3.2. The sub-sections below provide the procedure to enter/edit the other characteristics of Multiplexers.

Note on Sample encoding using MIDs and Multiplexers:

In the standard non-MID demultiplexing scheme, the AVA software looks for the template-specific primer sequences (Primer 1 and Primer 2) of the defined Amplicons at the beginning of each read. Once the Amplicon to which a read belongs is identified, the Sample-Amplicon associations defined for the Read Data Set that the read comes from are used to assign the read to its appropriate Sample. In other words, when MIDs are not used, the assignment of a read to an Amplicon, using the template-specific primers, is sufficient to further assign the read to the proper Sample. As explained before (see section 1.1.1.6, and Note and Caution sidebars in section 1.3.1.2), this scheme imposes the restriction that an Amplicon may only belong to a single Sample within a Read Data Set, to allow for unambiguous Sample assignment of the reads.

MIDs and Multiplexers allow this restriction to be lifted and experiments to be designed in which reads from a given Amplicon are monitored in multiple Samples, in the same Read Data Set. In this scheme, the MID sequence detected within a read is used to assign reads to Samples. This MIDs to Samples assignment is the function of Multiplexers, as described in this section.

The Primer 1 and Primer 2 sequences of a read are still used, however, to determine to which Amplicon a read belongs. Moreover, different Amplicons, within a Read Data Set, may be associated with different Multiplexers. Thus, when MIDs and Multiplexers are used, the demultiplexing process involves:

1.	decoding the Primer1 and Primer2 regions of the read to determine which Amplicon it represents;

2.	using the user-specified association between Amplicons and Multiplexers for the Read Data Set to determine the appropriate Multiplexer to apply; and

3.	using the MID sequence detected in the reads, in conjunction with that Multiplexer, to assign the read to the proper Sample.

1.3.2.7.1

To Enter or Edit the Sample Encoding using Multiplexers

The AVA software provides 4 ways to encode the Sample to which a read belongs in the Multiplexer, based on the construction of the libraries (see section 4.6 for details on Amplicon library design with MIDs). The proper option must be selected from a drop down menu in the Multiplexers Definition Table (Figure 1‑33). The options, further described in the sub-sections below, are:

•

Both

•

Either

•	Primer 1 MID

•	Primer 2 MID

Figure 1‑33: The Encoding drop down menu, on the Multiplexers tab

Selecting the proper encoding: It is crucially important to select the encoding method that truly corresponds to the way the libraries were prepared. For example, if libraries were prepared with ‘Either’ chemistry in mind, it may be tempting to use a ‘Primer 1 MID’ or ‘Primer2 MID’ encoded Multiplexer since the distal MID gets discounted in favor of the proximal MID in ‘Either’ encoding. However, the AVA software needs to know that MIDs are expected to be found at both ends: without that knowledge, the trimmer might get a suboptimal alignment of the distal primer, which in certain cases could drop valid reads out of the analysis.

1.3.2.7.1.1

“Primer 1 MID” and “Primer 2 MID” Encoding

The “Primer 1 MID” and “Primer 2 MID” encoding options assume that the libraries were prepared with MID sequences incorporated in the design of only one of the Adaptors used in the preparation of the Amplicon libraries. The MID is placed between the sequencing key and the template-specific primer that will be identified either as Primer 1 or as Primer 2 (as entered in the Amplicon definition table).

When either of these encoding options is selected for a Multiplexer, only the corresponding Primer MID field, Primer 1 MIDs or Primer 2 MIDs, needs to be filled in the Multiplexer’s Definition Table, to identify the MIDs used in the scheme (see section 1.3.2.7.2). For example, a Multiplexer encoded as “Primer 1 MID” will have an empty column in the Definition Table for the “Primer 2 MIDs” field. The maximum number of Samples that can be encoded with this scheme is equal to the number of MIDs defined in the Primer 1 MIDs or Primer 2 MIDs field.

The AVA software uses the encoding type to automatically determine where to search for MIDs within reads, taking read orientation into account. For example, if both forward and reverse reads are sequenced for an experiment where the “Primer 1 MID” encoding is being used, forward reads will have the MID at the beginning of the read, just before of the template-specific Primer 1 sequence; and reverse reads will have the reverse complement of the MID near the end of the read, just after the reverse complement of the template-specific Primer 1 sequence. For this reason, if reads will be obtained in both orientations (as is recommended) and Primer 1 MID or Primer 2 MID encoding is used, it is important to design Amplicons of a length shorter than the read length provided by the sequencing Run. If the ability to read through the Amplicons is in doubt, the “Either” encoding (section 1.3.2.7.1.3) should be used instead, to guarantee that both forward and reverse reads have an MID at their beginning.

For details on Sample assignment using the Primer 1 MID or Primer 2 MID encoding options, see section 1.3.2.7.3.1.

1.3.2.7.1.2

“Both” Encoding

”Both” encoding involves the incorporation of MID sequences in both Adaptors used in the preparation of the Amplicon libraries, such that each read contains both a Primer 1 MID and a Primer 2 MID. Therefore, both the “Primer 1 MIDs” and the “Primer 2 MIDs” fields of the Multiplexer Definition Table must have at least one MID selection (see section 1.3.2.7.2).

With this encoding scheme, both a Primer 1 MID and a Primer 2 MID must be observed in a read to assign it to the proper Sample. Note that there is no requirement that the same set of MIDs be used at both ends of the Amplicons: the two MIDs used to determine Sample assignment are completely independent from one another. In addition, with the “Both” encoding, the order of the appearance of MIDs is significant: for example, “Primer 1 Mid1 – Primer 2 Mid6” is a different encoding than “Primer 1 Mid6 – Primer 2 Mid1”. Given this combinatorial feature, the maximum number of Samples that can be encoded with this scheme is equal to the product of the number of MIDs defined in the Primer 1 MIDs and the Primer 2 MIDs fields.

It is also important to remember that in order to be able to read the distal MID, the length of the Amplicon library product must be within the read length provided of the sequencing Run script. For details on Sample assignment using the Both encoding option, see section 1.3.2.7.3.2.

1.3.2.7.1.3

“Either” Encoding

The “Either” encoding method is a hybrid between the single primer MID and “Both” methods. The libraries are tagged with MIDs on both the Primer 1 and Primer 2 ends, but only one MID needs to be observed on a read to assign it to the proper Sample. This can be useful if the Amplicon library products are longer than the read length of the sequencing Run, and you are sequencing them from both ends. One limitation of this encoding scheme is that a given MID can be used for only one Sample, for each of the Primer 1 and Primer 2 ends.

As with the Both encoding scheme, there is no requirement that the same set of MIDs be used at both ends of the Amplicons. Although the number of MIDs used at the Primer 1 and Primer 2 ends will typically be the same, the software even allows degenerate designs in which the “Either” encoding is used and the number of Primer 1 MIDs and Primer 2 MIDs differ. Regardless, a read must be able to be uniquely assigned to the same Sample whether its Primer 1 MID or the Primer 2 MID are used for demultiplexing. Therefore, the maximum number of Samples that can be encoded with a Multiplexer that uses this scheme is equal to the smaller of the number of MIDs defined in the Primer 1 MIDs and Primer 2 MIDs field.

For details on Sample assignment using the Either encoding option, see section 1.3.2.7.3.3.

1.3.2.7.2

To Enter or Edit the Primer 1 MIDs and Primer 2 MIDs

The user must specify the list of MIDs that the AVA software must search for to demultiplex reads using a Multiplexer. This information is set in the Primer 1 MIDs and the Primer 2 MIDs columns of the Multiplexer Definition Table. If Primer 1 MID or Primer 2 MID encoding is chosen, only the corresponding ‘Primer MIDs’ cell is available for that Multiplexer; if Either or Both encoding is chosen, both cells are available and must be filled.

To specify the MIDs for one end of a Multiplexer, double-click on the appropriate ‘Primer MIDs’ cell for that Multiplexer. The Edit Primer 1 MIDs (or Edit Primer 2 MIDs) window opens (Figure 1‑34). The window will not open unless at least one MID entry has already been specified into the MID Definition Table (though the MIDs do not have to have sequences defined at this stage). Select the MIDs of interest on the list on the left, and click

. To remove MIDs that have been previously selected, highlight them in the list on the right, and click

Certain shortcuts are available to carry out these tasks, such as an

and a

button; and an MID Group drop down menu allows the user to restrict the list on the left to the MIDs contained in any of the MID Groups available in the Project. Also, if the Primer 1 and Primer 2 MIDs are the same, you can first define the Primer 1 MIDs and then use the

button in the Edit Primer 2 MIDs window (Figure 1‑34B), to create the same list for that set.

Figure 1‑34: (A) The Edit Primer 1 MIDs window and (B) the Edit Primer 1 MIDs window. Note the

button in the Edit Primer 2 MIDs window.

The MID Group drop down menu also contains some “virtual” groups that are automatically generated by the AVA software based on the MIDs currently defined in the Project. Figure 1‑35 shows an example where all 14 of the 454Standard MIDs (10-mers) have already been loaded into the Project, and four new MIDs have been added without groups: two 6-base MIDs (Mid15 and Mid16) and two MIDs for which no sequence has yet been defined (Mid17 and Mid18).

Figure 1‑35: MID Tree and Definition Table view showing the 454Standard group MIDs, plus four newly defined MIDs (Mid15 - Mid18), two 6-mers and two that have no defined sequence.

Figure 1‑36A, then shows the MID Group drop down menu for the MIDs in Figure 1‑35:

•	The AVA software always provides an “[All MIDs]” option on this menu to allow all the MIDs defined in the Project to be viewed in the left list, irrespective of their group status (and even if an MID in the Project has not yet been assigned a sequence).

•	In addition, the software creates virtual MID Groups based on the length of the MIDs defined in the Project. This is useful because, as mentioned above (see Note in section 1.3.2.6), all the MIDs used on a given end of an Amplicon must be of the same length.

◦

Note that MIDs without a defined sequence will appear in all length-restricted lists (e.g. see Figure 1‑36B). This allows undefined MIDs to be selected in a Multiplexer scheme and defined later. Once an MID has a sequence defined, it will lose its wild card status and will only appear in the list appropriate to its length.

Figure 1‑36: (A) The Edit MIDs window showing the MID Group drop down menu. A defined group “454Standard” is listed along with three custom automatically generated groups, the “[All MIDs]” group and two groups based on MID length. (B) The “[Length 6 compatible MIDs]” group restricts the left list to those MIDs that are exactly 6 bases long, along with those that have no sequence yet defined.

The lower part of the Edit Primer 1 MIDs (or Edit Primer 2 MIDs) window provides information (and errors or warnings, as appropriate) concerning the MIDs selected. For example, this includes a summary of the number of MIDs selected, their length, and the minimum edit distance (the minimum number of insertion, deletion, or substitution sequencing errors that could turn one of the selected MID sequences into one of the others) (see Figure 1‑34, above).

The types of errors and warnings provided may include MIDs not all the same length, or undefined MIDs (Figure 1‑37). Note that the software gives the benefit of the doubt to undefined MIDs, and calls the attention of the user with a warning but does not assume an error. This provides the advantage that the structure of a Multiplexer can be defined independently, and possibly in advance, of the knowledge of the MID sequences themselves. However, prior to computation, all the MIDs used in defining Multiplexers that are associated with active Read Data Set must, naturally, be defined. The software also calculates the minimum edit distance even for defined MIDs of different lengths, assuming that corrections will be made prior to Project computation (i.e. that MIDs of unequal length will be corrected or eliminated).

Figure 1‑37: Examples of errors and warnings flagged on the Edit Primer 1 MIDs window.

1.3.2.7.3

To Enter or Edit the Samples Assignment

The most complex part of setting up a Multiplexer is to specify the assignment of the MIDs to the Samples. This is done in the Edit Samples window, which is accessed by double-clicking in the cell of the Samples column, for the Multiplexer of interest (in the Multiplexer Definition Table). This window can take 3 different forms, depending on the type of encoding selected, as described in the sections below. Note that Primer 1 MIDs and/or Primer 2 MIDs, as appropriate, must have been selected for that Multiplexer for the Edit Samples window to be available; if any errors or warnings exist regarding the selected MIDs (see section 1.3.2.7.2), these will be displayed in the Edit Samples window as well.

1.3.2.7.3.1

Sample Assignment with “Primer 1 MID” or “Primer 2 MID” Encoding

With these single-end MID encoding schemes, the Edit Samples window simply lists all the MIDs selected for the Multiplexer (see section 1.3.2.7.2), and the Sample is selected from the drop down menu (or can be typed) in the cell to the right of each MID name (Figure 1‑38A). The user can also type into the cells the names of Samples that have not yet been defined in the project: new samples with those names will automatically be created and appear in the Project’s Samples Definition Table when the user clicks

. These samples will not be created, however, if the user clicks

. A Sample assignment may be removed from a given MID by choosing the “--remove‑‑“ option from the drop down menu.

If the drop down menus are too narrow to display the full Sample names, you can widen the Edit Samples window, making more room for the ‘Sample’ column.

Certain shortcuts are available on this window as well: clicking the

button assigns default-named Samples to any MID that does not yet have an assigned Sample (Figure 1‑38B); and a

button empties all the Sample cells. The default Sample names contain three parts, in the following format:

Sample_<Multiplexer name>_<MID name>

As with typed-in novel Sample names, the Autofill Samples will only be added to the Project if the user clicks the

button.

Figure 1‑38: The Edit Samples window, for Primer 1 MIDs encoding.

A functional Multiplexer must specify at least one Sample assignment, but it is not formally necessary to fill all cells of the Table. This can be useful if a subset of the selected MIDs have not been used in the experiment, but were known to have been used in a previous experiment: specifying them allows the system to search for these MIDs as potential contaminants and prevent them from being misinterpreted as an erroneous version of one of the MIDs actually used in the Project to demultiplex Samples. On the other hand, it is possible to assign the same Sample to multiple MIDs (but not the opposite). Since this could be a legitimate experimental set up, it does not elicit an error message; however, a warning is displayed to draw the user’s attention to this unusual assignment (Figure 1‑38B).

In a manner analogous to the Edit Primer 1 MIDs or Edit Primer 2 MIDs windows (section 1.3.2.7.2), a summary of the assignment scheme is provided at the bottom of the Edit Samples window, including information on the number of MID-Sample associations defined and the total number that can be defined with the MIDs selected (Figure 1‑38). Any errors and warnings associated with the MIDs are also shown here, to alert the user that action must be taken to complete or correct the MID definitions or the Sample assignments (Figure 1‑39).

Figure 1‑39: The Edit Samples window for Primer 1 MID encoding, showing one error and two warnings regarding either the MIDs currently selected for this Multiplexer, or the Sample assignments. The MIDs are listed in sorted order based on their MID Group (not displayed, but visible in the tooltip that will appear if the user hovers the mouse over the MID name) and then their MID name. MIDs without an associated MID Group appear before those with an MID Group. In this example, Mid16 and Mid17 were not assigned to an MID Group and so appear before Mid1. Had they been part of the same MID group as Mid1 they would have appeared later in the list, as expected.

1.3.2.7.3.2

Sample Assignment with “Both” Encoding

With the “Both” encoding scheme, two MIDs must be specified for each Sample, one attached to Primer 1 and one to Primer 2. In this case, therefore, the Edit Samples window displays a two-dimensional Table where the MIDs selected for the Primer 1 side occupy one dimension (rows or columns) and those for Primer 2 occupy the other (Figure 1‑40). In a manner analogous to the single-MID Sample encoding seen above, Sample assignment is done by selecting the Sample name from the drop down menu (or by typing it) in the cell at the intersection of the two encoding MID names.

If the drop down menus are too narrow to display the full Sample names, you can widen the Edit Samples window, making more room for the ‘Sample’ columns. You can also resize individual columns by dragging on the column header separators to the left or right, until the column of interest has the proper width to allow the display of the full Sample names.

The other features of this window (can have empty cells, shortcut buttons, summary and error/warning reporting, etc.) are the same as for the Primer 1 MID or Primer 2 MID encoding, described above. For “Both” encoding, the names of the Autofill Samples are of the form:

Sample_<Multiplexer name>_<Primer 1 MID name>_<Primer 2 MID name>

It is important to be aware of the directionality of the Amplicons when assigning the Samples to MID pairs: MIDs are selected separately for the Primer 1 and Primer 2 sides to support this directionality. In the Edit Samples window, the side corresponding to the two selected MID sets are identified by a “1” and a “2” with arrowheads, on the top-left corner of the Table. This is illustrated on Figure 1‑40: in this example, the table of the Edit Samples window has been populated using the

button; the Primer 1 – MID1: Primer 2 – MID2 pair encodes Sample Sample_Multi7_Mid1_Mid2, while the Primer 1 – MID2: Primer 2 – MID1 pair encodes Sample Sample_Multi7_Mid2_Mid1. For convenience, a

button can transpose the table.

Figure 1‑40: The Edit Samples window for the “Both” encoding scheme, showing Samples that were assigned using the

button.

1.3.2.7.3.3

Sample Assignment with “Either” Encoding

With the “Either” encoding, Amplicons also have two MIDs so the Edit Samples Table is two-dimensional as well. However, contrary to the situation with Both encoding, the MID from only one end is used to assign any given read to the proper Sample. This situation is akin to the Primer 1 MID and Primer 2 MID encoding cases, and likewise, each MID (at a given end) can encode only one Sample.

To help in this, the software grays out cells that become ineligible as Samples are assigned to Primer 1 MID – Primer 2 MID pairs. In the simplest case, the libraries are designed such that the same MIDs are placed at both ends of each Amplicon (Figure 1‑41A). For an Either encoded Multiplexer, the

function is only enabled for this type of symmetric design (AutoFill expects to make sample assignments along the diagonal where the same MID is used on each end of the read: Mid1-Mid1, Mid2-Mid2, etc.).

However, asymmetric designs are also legitimate. The software flags this with a warning in case the asymmetry was unintended (Figure 1‑41B). Even if the same set of MIDs are selected for both the Primer 1 MIDs and the Primer 2 MIDs series (a symmetrical design), the Sample assignment does not have to be along the diagonal in the grid (Mid1-Mid1, Mid2-Mid2, etc.) as it would be with an AutoFill. As long as no MID at either end is assigned to more than one Sample, and every MID on one side that has a Sample assignment has some corresponding MID on the other side with the same Sample assignment, the design is still valid. Again, mis-assignment is prevented by graying out the ineligible cells (Figure 1‑41C).

A	B
C

Figure 1‑41: The Edit Samples window, for Either encoding. (A) Symmetrical design with Sample assignment down the diagonal. (B) Asymmetrical design with different sets of Primer1 MIDs and Primer 2 MIDs and a warning about the asymmetry. (C) Symmetrical design but with a Sample assignment deviating from the diagonal.

In fact, it is even possible to have a different number of MIDs selected on the Primer 1 and Primer 2 sides. When this kind of design is used, the software displays a warning that there are unequal numbers of Primer 1 and Primer 2 MIDs, and specifies the number of unbalanced associations (Figure 1‑42). In this special case, one or more MIDs will have to be used more than once, yet the constraint that a given MID at a given end of the Amplicons must specify a single Sample (to allow for unambiguous assignment of the reads) must be respected. To accomplish this, the AVA software restricts the Sample choices in cells that may receive such secondary assignments (highlighted with a thicker gray border), to Samples already specified for a Primer 1 MID or a Primer 2 MID. Some of the specific circumstances one might encounter are illustrated in Figure 1‑42.

A	B
C	D

Figure 1‑42: Edit Samples window for an Either-encoded Multiplexer with an unbalanced design. There are 4 Primer 1 MIDs and 3 Primer 2 MIDs. (A) With two sample assignments already made, the only cells that are not constrained at all (Mid3-Mid3 and Mid4-Mid3) are shown as totally white. Cells that have a thicker gray border are available for selection but are constrained to a single choice by previous Sample assignments. (B) The Mid3-Mid3 cell has an unconstrained Sample list and Sample 3A is being selected. (C) After the Sample assignment, the Mid4 row consists of three constrained cells with thick gray borders. The Mid4-Mid1 cell only allows Sample 1A to be selected because the Primer 2 MID Mid1 is already being used to encode that Sample. (D) Similarly, The Mid4-Mid2 cell only allows Sample 2A to be selected because the Primer2 MID Mid2 is already being used to encode Sample 2A.

Again the other features of the Edit Samples window for “Either” encoding (can have empty cells, shortcut buttons, summary and error/warning reporting, etc.), are the same as for the Primer 1 MID, Primer 2 MID, or Both encoding, described above. Autofill Samples are handled the same way as for Both encoding.

As above, if the drop down menus are too narrow to display the full Sample names, you can widen the Edit Samples window, making more room for the ‘Sample’ columns. You can also resize individual columns by dragging on the column header separators to the left or right, until the column of interest has the proper width to allow the display of the full Sample names.

1.3.2.7.4

Using Multiplexers for more than one Read Data

Once a Multiplexer has been created and defined, it can be used on any number of Read Data Sets in the Project, as long as these Read Data Sets share the same MIDs, encoding, and MID-to-Sample associations. Furthermore, each Multiplexer-Read Data Set pairing can even be used to demultiplex distinct sets of Amplicons, as defined by the Sample-Amplicon-Read Data Set triad associations (Figure 1‑43).

Figure 1‑43: A single Multiplexer (Multi_01) is associated with 4 different Read Data Sets in the Read Data Tree. In the context of the first two Read Data Sets, the same set of Amplicons is being measured (amp1-amp4), but different Amplicons are being measured for each of the remaining Read Data Sets (amp5-amp8 for the third Read Data Set and amp9-amp12 for the fourth Read Data Set).

The AVA software also allows for Multiplexers to be duplicated. Although it is not necessary to define multiple exact copies of Multiplexers within a Project, as just discussed, duplication may be useful if multiple Multiplexers need to be defined, that share common baseline features such as the encoding scheme and specific MIDs on each side of their Amplicons. This is done using the “Duplicate item” button on the left margin of the Project Tab (see section 1.3.2): a copy of a Multiplexer created this way retains the encoding and MID settings of the original.

The “Select Amplicons associated with item” button can also provide a very useful shortcut when a given set of Amplicons is to be measured by multiple Read Data Set – Multiplexer pairs. This button is also located on the left margin of the Project Tab, and its functionality is described in section 1.3.1. Selecting a large number of disparate Amplicons from the Amplicons Definition Table, to associate them to a Multiplexer, can be laborious and painstaking; if many Multiplexers require the same (or similar) Amplicon associations, you need to create only the first of these Multiplexers manually, then select it in the Read Data Tree and click the “Select Amplicons associated with item” button; the software will switch to the Amplicons Definition Table sub-tab, and the subset of the Amplicons that are associated with the original Multiplexer will be selected, ready to be dragged to another Multiplexer in the Tree.