Harnessing the 8- and 12-multichannel pipette
Liquid handling is a challenge for multiplex designs, particularly for large designs with a large number of replicates. For example, loading the XL5 requires 1,355 liquid transfers.
One way to speed this process is to use multichannel pipettes. These pipettes have
8 or 12 tips in a row, and as such can pull an equal volume from a whole row or column of a 96 well plate.
One way to use these multichannel pipettes is to carry out 8 or 12 parallel designs. For example, using an 8 channel pipette, we could treat each of the 12 columns (1-12) of a plate as a well, or if we spread the multiplexed design across two plates we could have 24. Similarly, using a 12 channel pipette, we could use each row (A-H) on two plates as a virtual 16 well assay run in parallel 12 times.
Pipette configuration |
Number of plates |
Number of "wells" |
Max samples (replication) |
Liquid handling transfers |
Total samples run |
8 channel |
1 |
12 |
66 (x2) |
132 |
528 |
|
|
|
20 (x3) |
60 |
240 |
|
2 |
24 |
276 (x2) |
552 |
2208 |
|
|
|
82 (x3) |
246 |
656 |
12 channel |
2 |
16 |
120 (x2) |
240 |
1440 |
|
|
|
37 (x3) |
111 |
444 |
|
3 |
24 |
276 (x2) |
552 |
3312 |
|
|
|
82 (x3) |
246 |
984 |
While these smaller designs achieve a compression of 1.7x (with 3 fold replication) up to 11.5x (with 2 fold replication), they also achive the benefits of 8 and 12 parallel liquid transfers.
8-channel rotations and inversion
Another strategy for exploiting the geometry of a 96 well plate and an 8-channel pipette is to use the fulller space of rotations and inversions of the pipette itself.
To illustrate this, consider an 8x8 grid of wells on the edge of a 96 well plate:
| A | B | C | D | E | F | G | H |
1 | | | | | | | | |
2 | | | | | | | | |
3 | | | | | | | | |
4 | | | | | | | | |
5 | | | | | | | | |
6 | | | | | | | | |
7 | | | | | | | | |
8 | | | | | | | | |
To add a line of samples (#1-#8) to the plate, there are 4 distinct patterns possible:
(1) Vertical - up
| A | B | C | D | E | F | G | H |
1 | #1 | | | | | | | |
2 | #2 | | | | | | | |
3 | #3 | | | | | | | |
4 | #4 | | | | | | | |
5 | #5 | | | | | | | |
6 | #6 | | | | | | | |
7 | #7 | | | | | | | |
8 | #8 | | | | | | | |
|
(2) Vertical - down
| A | B | C | D | E | F | G | H |
1 | #8 | | | | | | | |
2 | #7 | | | | | | | |
3 | #6 | | | | | | | |
4 | #5 | | | | | | | |
5 | #4 | | | | | | | |
6 | #3 | | | | | | | |
7 | #2 | | | | | | | |
8 | #1 | | | | | | | |
|
(3) Horizontal - left
| A | B | C | D | E | F | G | H |
1 | #1 | #2 | #3 | #4 | #5 | #6 | #7 | #8 |
2 | | | | | | | | |
3 | | | | | | | | |
4 | | | | | | | | |
5 | | | | | | | | |
6 | | | | | | | | |
7 | | | | | | | | |
8 | | | | | | | | |
|
(4) Horizontal - right
| A | B | C | D | E | F | G | H |
1 | #8 | #7 | #6 | #5 | #4 | #3 | #2 | #1 |
2 | | | | | | | | |
3 | | | | | | | | |
4 | | | | | | | | |
5 | | | | | | | | |
6 | | | | | | | | |
7 | | | | | | | | |
8 | | | | | | | | |
|
Using these four variants allows us to use more of the coding bandwith of the 64 assays. For example, the combinations of the two vertical options makes the 8 wells across behave like 16 wells. Similarly, the 8 wells down with left/right inversions behaves like 16 wells. Using these all four varints together, we should have access to the equivalent of 16+16 = 32 wells total. This is still half of the maximum 64 wells, but is accomplished with 1/8th of the pipetting steps.
If we are careful, we can blend horizontal with vertical patterns too, but only if they have oppoiste orientation. For exampe, we
can not use vertical-up with horizontal-left because this will result in compound #1 being tested twice with iteslf in well A1. However, we can blend vertical-up with horizontal-right to produce designs that don't produce self-duplicating wells.