Arichard:Notebook/2013/August

From ZhangLabWiki
Jump to navigation Jump to search

August, 2013[edit]

August 19, 2013[edit]

Can I implement Quartz WTA in PDMS microwell arrays?[edit]

  • Loading is a problem. For MIDAS, we load 3 ul of 10-15 cells/ul in PBS over each array. Quartz is written for 1 cell in 0.4 ul 0.5% NP-40 in a 0.2 ul low bind tube.
    • I don't have to load the entire array. I could load 0.4 ul of 10-15 cells/ul into the center to avoid changing concentrations.
  • Quartz sorts into the lysis buffer, but I want to load and image the cells in the wells before starting.
    • Imaging whole cells means that my loading buffer has to be isotonic.
  • Lysing is also a problem.
    • MIDAS lyses with 1.5X volumes of ALS, neutralized by the same volume of NS --> 4X volume after lysis.
    • Quartz lyses in 0.4 ul of 0.5% NP-40. This is also hypotonic, which helps lysis.
  • I could load and image in PBS, then add enough NP-40 to make 0.5%.
    • Mixing would be critical, as in MIDAS ALS/NS.
    • Even with 0.5% NP-40, the cells may not lyse efficiently in an isotonic solution.
  • The best solution may be to use the MIDAS loading and lysis protocol with scaled down volume.
    • Load 0.4 ul.
    • Lyse with 0.6 ul ALS
    • Neutralize with 0.6 ul NS.
    • Total volume = 1.6 ul.

August 21[edit]

MIDAS diffusion testing[edit]

Motivation: MIDAS relies on a digital amplification profile in the microwell array, with approximately 1 cell per 10 wells. This helps to ensure that occupied wells contain only single cells. We also tend to pick positive wells surrounded by negative wells. These measures give us reasonable confidence that data generated will be from single cells. This assumption requires, however, that DNA from each lysed cell remains in its well, fluidically isolated from other wells.

Objective: Test for diffusion of DNA out of an occupied well into surrounding wells.

Outline: We ran MDA overnight per MIDAS protocol, and extracted amplicons the next day for 4 samples. In addition, however, for each positive well we picked three adjacent, negative wells to act as no template controls. MDA is very sensitive, and even 1 kb fragments which had diffused into neighboring wells and amplified should be detected by a second round of qMDA.

Conclusion: All of our samples were at or below 10 pg relative to the ladder. We expect close to 1 ng after overnight MDA. Perhaps qMDA is not efficiently detecting first round MDA products, or first round MDA is being inhibited. Alternatively, we may be losing product at some step.

2013_08_21 qmda

August 22[edit]

Pipet shearing test by qMDA[edit]

Motivation: Several qMDA tests have failed to detect extracted amplicons above a background contamination level of ~1 fg.

  • We expect ~1 ng from overnight microwell MDA.
  • All ladders (10 ng, 1 ng, 100 pg, 10 pg) were treated the same as samples after extraction. All ladder amounts showed expected CT values.
  • We suspect that qMDA is inefficient on small templates (hyperbranch structure forms inefficiently), and that the glass pipets are shearing the sample.

Objective: Test whether shearing the template through glass pipets inhibits qMDA. We have three conditions:

  • No shear (ladder).
  • Shear 1X (pipet up and down), 2 bio replicates.
  • Shear 5X, 2 bio replicates.

Denaturation and neutralization in 2X volume, split into white strip tubes and add master mix.

  • 8 samples, in duplicate (16 total):
    • 10 ng
    • 1 ng
    • 100 pg
    • 10 pg
    • 1 ng, 1X shear, rep 1
    • 1 ng, 1X shear, rep 2
    • 1 ng, 5X shear, rep 1
    • 1 ng, 5X shear, rep 2

Protocol

  1. 2 ul sample plus 3 ul ALS, 3 min @ RT.
  2. On ice, add 3 ul NS --> 8 ul.
  3. Split, 4 ul and 4 ul in white tubes.
  4. Add 16 ul master mix.
  • Master mix (18X):
    • 36 ul Phi29 buffer.
    • 18 ul thio N6.
    • 14.4 ul RepliPhi dNTP.
    • 2.88 50X SYBR.
    • 18 ul Phi29.
    • 198.72 ul H2O.

30 degC, measure every 6 min, up to 99 cycles (or completion).

Conclusion: No difference between 1 ng ladder, 1 ng 1X shear, and 1 ng 5X shear. Pipet shearing does not appear to affect efficiency of qMDA.

2013_08_22 qmda shear test


August 23[edit]

Oil and BSA incubation time test by qMDA[edit]

Motivation: Same as August 22. We need to find out why qMDA does not detect even the input amount, even after MDA, in microwells. Loss may be up to 90%.

  • Shearing due to glass pipets has been ruled out.
  • Shearing test was done on glass slides. We need to test PDMS arrays.
  • One possibility is that DNA is being removed with the oil.
  • Another is that DNA is binding to the well (in a time dependent manner), despite BSA coating.

Objective: Test oil removal and DNA binding time as factors inhibiting detection by qMDA after extraction from PDMS microwell arrays.

  • 4 conditions:
    • Load, extract immediately.
    • Load, add 75 ul oil overlay, remove oil, extract immediately.
    • Load, wait 1 hr, extract.
    • Load, add 75 ul oil overlay, wait 1 hr, remove oil, extract.

All conditions were loaded with 10 ul of 10 ng/ul genomic E. coli DNA.

Conclusuion: 1 hr, no oil condition evaporated. Should have used coverslip. No oil, no wait condition had an excessive amount of DNA solution on top of the wells, which led to extraction volumes greater than 20 nanoliters (well volume) and lower CT values. The oil conditions did not show a difference between 1 hour and no wait. The DNA may be binding, but just very quickly. The wells still showed large loss of DNA.

Note: 200 pg and 20 pg are on top of each other in the ladder.

2013_08_23 qmda oil and time test

August 24[edit]

qMDA BSA test[edit]

Motivation: We still need to solve the problem of missing DNA after extraction from the microwell arrays.

  • If an array is loaded with 10 ng/ul DNA, each 20 nl well should contain 200 pg. We are seeing much less than that by qMDA.
  • Many possibilities have been ruled out.
    • Not due to shearing (8/22).
    • Not due to DNA binding, time dependent (8/23).
    • Not due to oil removal (8/23).
  • We now hypothesize that the BSA coating is the problem.

Objective: Test whether BSA coating time affects PDMS microwell array retention of DNA. Secondary: Test whether a small amount of detergent improves recovery. Note: We have also gone back to using 10 ul of 1% BSA; Larger volumes are easier to spread over the array. We had been using 5 ul for several weeks.

  • 4 conditions:
    • No BSA
    • 30 min BSA
    • 60 min BSA
    • 60 min BSA + Tween
      • DNA loaded also contained Tween

2013_08_27 qmda BSA time and tween

Conclusion: 30 min BSA condition showed less recovery than 60 min. No BSA showed even less. No significant difference between 60 min BSA with or without Tween.