Matt:LabNotes/2014-3-10: Difference between revisions
Jump to navigation
Jump to search
>Mzcai m (→Secondary RCA) |
>Mzcai |
||
(6 intermediate revisions by the same user not shown) | |||
Line 6: | Line 6: | ||
*Try to generate highly multiplexed 2nd rolonies with 12k padlock probeset | *Try to generate highly multiplexed 2nd rolonies with 12k padlock probeset | ||
**30.3 pmole of Agi26k0gap prepared [[Matt:LabNotes/2014-3-5|LabNotes/2014-3-5]] | **30.3 pmole of Agi26k0gap prepared [[Matt:LabNotes/2014-3-5|LabNotes/2014-3-5]] | ||
***30.3 pmole / 200ul = 0.1515 pmole/ul = 0.1515 umole/L = 151. | ***30.3 pmole / 200ul = 0.1515 pmole/ul = 0.1515 umole/L = 151.5nM | ||
****151. | ****151.5nM/12,000 probes = 0.0126 nM = 12.6 pM | ||
===Procedure=== | ===Procedure=== | ||
*Using Mattek dish #2 from [[Hosuk:LabNotes/2014-3-3]] that was fixed in 10% formalin | *Using Mattek dish #2 from [[Hosuk:LabNotes/2014-3-3]] that was fixed in 10% formalin | ||
====Padlock Probe Capture==== | |||
#Strip off dye-probes Hosuk used with 80% formamide pre-heated to 75C and incubated @45C for 15min | #Strip off dye-probes Hosuk used with 80% formamide pre-heated to 75C and incubated @45C for 15min | ||
#Wash with 1X PBS twice | #Wash with 1X PBS twice | ||
Line 56: | Line 57: | ||
#Wash 1X PBS twice | #Wash 1X PBS twice | ||
==== | ===Detection=== | ||
* | ====Initial Imaging==== | ||
*3/12/2014 | |||
*Check to see if 2nd rolony generation was successful by detecting a large fraction | |||
**Use 1uM dcProbe0-Cy3 and 1uM dcProbe0-FAM in 30% formamide + 2X SSC | |||
***This should detect ~1/2 of all 2nd rolonies | |||
===Results=== | ===Results=== | ||
*Using PISA7 Paramters: Gaussian std (3), upper bound (-2e-4), area upper (50), area lower (4), axratio lower (.6), circ upper (1.6), circ lower (.8), perim conn (8), bkgmult lower ( | ====Initial Imaging==== | ||
*Using PISA7 Paramters: Gaussian std (3), upper bound (-2e-4), area upper (50), area lower (4), axratio lower (.6), circ upper (1.6), circ lower (.8), perim conn (8), bkgmult lower (variable) | |||
*First used PISA to examine the images with varying bkgmult lower (range: 2-9) | |||
{| {{table}} | |||
| align="center" style="background:#f0f0f0;"|'''''' | |||
| align="center" style="background:#f0f0f0;"|'''Bkgmult''' | |||
| align="center" style="background:#f0f0f0;"|'''Rolony Count''' | |||
| align="center" style="background:#f0f0f0;"|'''Notes''' | |||
|- | |||
| NoDye_G680_Cy3||7||116|| | |||
|- | |||
| NoDye_G680_Cy3||9||31|| | |||
|- | |||
| NoDye_L2%_G680_FAM||7||202||picked up mostly nuclei | |||
|- | |||
| NoDye_L2%_G680_FAM||9||7|| | |||
|- | |||
| dcProbe0-Cy3_G550||7||178|| | |||
|- | |||
| dcProbe0-Cy3_G550||9||162|| | |||
|- | |||
| dcProbe0-Cy3_G550||5||208|| | |||
|- | |||
| dcProbe0-Cy3_G550||3||265||decreasing bkgmult detects more "true" features without significantly counting artifacts | |||
|- | |||
| dcProbe0-Cy3_G550||2||278||doesn't detect anymore "true" features, just adds noise | |||
|- | |||
| dcProbe0-Cy3_G680||7||544|| | |||
|- | |||
| dcProbe0-Cy3_G680||9||489|| | |||
|- | |||
| dcProbe0-FAM_L5%_G550||7||129|| | |||
|- | |||
| dcProbe0-FAM_L5%_G550||3||209|| | |||
|- | |||
| dcProbe0-FAM_L5%_G550||2||245||still misses many while counting artifacts | |||
|- | |||
| dcProbe0-FAM_G550||7||52|| | |||
|- | |||
| dcProbe0-FAM_G550||5||68|| | |||
|- | |||
| dcProbe0-FAM_G550||3||108||even lowering it to 3 still won't count obvious features | |||
|- | |||
| dcProbe0-FAM_G600||7||313|| | |||
|- | |||
| dcProbe0-FAM_G600||5||364||good | |||
|- | |||
| dcProbe0-FAM_G680||7||519|| | |||
|- | |||
| dcProbe0-FAM_G680||9||375||not bad, pretty similar to G600 bkgmult5 | |||
|} | |||
'''Count Cells in Field of View''' | |||
*NoDye_L2%_G680_FAM had nuclei with high intensity fluorescence relative to background, which was really obvious after processing in PISA so counted nuclei in that image | |||
**110 Nuclei/Cells counted | |||
[[File:Agi26k0gap_PGP1f_InitialDetect_NucleiCount110.png|450px]] | |||
'''Best Images and PISA Settings''' | |||
{| {{table}} | |||
| align="center" style="background:#f0f0f0;"|'''''' | |||
| align="center" style="background:#f0f0f0;"|'''Image Name''' | |||
| align="center" style="background:#f0f0f0;"|'''Image''' | |||
| align="center" style="background:#f0f0f0;"|'''Feature Count''' | |||
|- | |||
| Cy3 Background||NoDye_G680_Cy3_Bkgmult9||[[File:NoDye_G680_Cy3_ch00_PISA7Overlay_Bkgmult9.png|450px]]||31 | |||
|- | |||
| FAM Background||NoDye_L2%_G680_FAM_Bkgmult9||[[File:NoDye_L2%25_G680_FAM_ch00_PISA7Overlay_Bkgmult9.png|450px]]||7 | |||
|- | |||
| dcProbe0-Cy3||dcProbe0Cy3_G550_Bkgmult3||[[File:DcProbe0Cy3_G550_Cy3_ch00_PISA7Overlay_Bkgmult3.png|450px]]||265 | |||
|- | |||
| dcProbe0-FAM||dcProbe0FAM_G600_Bkgmult5||[[File:DcProbe0FAM_600_FAM_ch00_PISA7Overlay_Bkgmult5.png|450px]]||364 | |||
|} | |||
'''Removing Background''' | |||
*'''dcProbe0-Cy3''' | |||
Ran PISA on NoDye_G680_Cy3_Bkgmult9 and dcProbe0Cy3_G550_Bkgmult3 | |||
Saved bw images in workspace as backgroundCy3 and dcProbe0Cy3 | |||
countOverlappedPoloniesv2(backgroundCy3,dcProbe0Cy3) | |||
ans = 0 | |||
comp_img1=zeros(xsize,ysize,3); | |||
comp_img1(:,:,1)=dcProbe0Cy3; | |||
comp_img1(:,:,2)=backgroundCy3; | |||
imwrite(comp_img1,'dcProbe0Cy3vsbackgroundCy3.png'); | |||
[[File:DcProbe0Cy3vsbackgroundCy3.png|450px]] | |||
*'''dcProbe0-FAM''' | |||
Ran PISA on NoDye_L2%_G680_FAM_Bkgmult9 and dcProbe0FAM_G600_Bkgmult5 | |||
Saved bw images in workspace as backgroundFAM and dcProbe0FAM | |||
countOverlappedPoloniesv2(backgroundFAM,dcProbe0FAM) | |||
ans = 0 | |||
comp_img1=zeros(xsize,ysize,3); | |||
comp_img1(:,:,1)=dcProbe0Cy3; | |||
comp_img1(:,:,2)=backgroundCy3; | |||
imwrite(comp_img1,'dcProbe0Cy3vsbackgroundCy3.png'); | |||
[[File:DcProbe0FAMvsbackgroundFAM.png|450px]] | |||
*Conclusion: background is not a factor | |||
'''Specificity of dcProbe0-Cy3 and dcProbe0-FAM''' | |||
*Is there overlap between the two? (There shouldn't be, they should be exclusive) | |||
countOverlappedPoloniesv2(dcProbe0Cy3,dcProbe0FAM) | |||
ans = 1 | |||
[[File:DcProbe0Cy3vsdcProbe0FAM.png|450px]] | |||
*The only overlap is in the bottom right corner and looks more likely due to nearby rolonies than non-specific binding of both dyes to some structure | |||
**Red is dcProbe0-Cy3 and Green is dcProbe0-FAM | |||
'''Overlay BrightField'''<br> | |||
[[File:DcProbe0Cy3vsdcProbe0FAM_BF.png|650px]] | |||
===Conclusion=== | ===Conclusion=== | ||
====Initial Imaging==== | |||
*Background signal did not play a factor since none overlapped with detected features during dcProbe0-Cy3 and dcProbe-FAM | |||
**We can say every feature we see is specific to 2nd rolony | |||
*dcProbe0-Cy3 and dcProbe0-FAM showed essentially no overlap, which gives us confidence these are true signals | |||
**Next time, we should be able to detect dcProbe1, dcProbe2, etc. and expect to see a fraction overlap with these | |||
*dcProbe0-Cy3 had 265 rolonies (~2.4/cell) | |||
*dcProbe0-FAM had 364 rolonies (~3.3/cell) | |||
*dcProbe0-Cy3(3641 probes) + dcProbe0-FAM (4097 probes) = 7,738 probes/12,964 total probes in Agi26k0gap = 60% | |||
**If these two dyes which can detect 60% of probes detected 629 2nd rolonies then we can approximate the total number of 2nd rolonies in this field of view = ~1050 | |||
*There are ~9.5 2nd rolonies/cell |
Latest revision as of 18:02, 14 March 2014
Agi26k0gap Secondary Rolonies Mattek Dish[edit]
- Since we now have a successful procedure for generating Secondary Rolonies
- Since we have optimized PGP1f sample prep by culturing without EGF (Hosuk:LabNotes/2014-3-3)
- In Harvard Science paper (Lee, J. et al), this culture condition resulted in more mRNA and less rRNA
- Also microarray data on GEO shows Agi26k0gap probe set targets many high expression transcripts in PGP1f cells Matt:LabNotes/2014-3-7
- Try to generate highly multiplexed 2nd rolonies with 12k padlock probeset
- 30.3 pmole of Agi26k0gap prepared LabNotes/2014-3-5
- 30.3 pmole / 200ul = 0.1515 pmole/ul = 0.1515 umole/L = 151.5nM
- 151.5nM/12,000 probes = 0.0126 nM = 12.6 pM
- 30.3 pmole / 200ul = 0.1515 pmole/ul = 0.1515 umole/L = 151.5nM
- 30.3 pmole of Agi26k0gap prepared LabNotes/2014-3-5
Procedure[edit]
- Using Mattek dish #2 from Hosuk:LabNotes/2014-3-3 that was fixed in 10% formalin
Padlock Probe Capture[edit]
- Strip off dye-probes Hosuk used with 80% formamide pre-heated to 75C and incubated @45C for 15min
- Wash with 1X PBS twice
- Anneal Padlock Probes @45C for 28hrs (Put in at 11:30a)
- 1X Ampligase Buffer
- ~12.6nM of each padlock probe
- 0.5 U/ul Ampligase
- Prepared everything on ice and then pre-heated to 75C for 5 min before adding
Component | Volume |
10X Ampligase Buffer | 20 |
30.3 pmole Agi26k0gap | 58 |
5U/ul Ampligase | 20 |
H2O | 102 |
Total | 200 |
Secondary RCA[edit]
- Aspirate but don't wash
- Pre-annealing the FISSEQ_ppRCA primer for the rolling circle amplification reaction.
- 1ul of RCA primer(200 uM, GATATCGGGAAGCTGA*A*G) in 199ul of 2xSSC/30% formamide
- Preheat primer at 60C (Tm = 55C when not in 30% formamide)
- Add 200ul and incubate at 45C for 15min.
- Aspirate, and wash using 2x SSC twice
- Aspirate, and wash using 0.1x SSC twice
- Prepare the rolling circle amplification reaction mixtures on ice.
- 171 ul ddH2O,
- 20 ul 10x Phi29 buffer,
- 2 ul 25 mM dNTP,
- 4 ul 2 mM aminoallyl dUTP
- 3 ul Phi29 DNA polymerase (Epicentre, low concentration)
- 200ul Total Volume
- Add and incubate 20hrs at 30C. (Put in at 4:15pm 3/11/14)
- Wash 1X PBS once
- Add 200ul BS(PEG)9 mix to each well and incubate @RT for 1hr
- Wash 1X PBS twice
- Add 200ul 1M Tris pH 8.0 @RT for 30min
- Wash 1X PBS twice
Detection[edit]
Initial Imaging[edit]
- 3/12/2014
- Check to see if 2nd rolony generation was successful by detecting a large fraction
- Use 1uM dcProbe0-Cy3 and 1uM dcProbe0-FAM in 30% formamide + 2X SSC
- This should detect ~1/2 of all 2nd rolonies
- Use 1uM dcProbe0-Cy3 and 1uM dcProbe0-FAM in 30% formamide + 2X SSC
Results[edit]
Initial Imaging[edit]
- Using PISA7 Paramters: Gaussian std (3), upper bound (-2e-4), area upper (50), area lower (4), axratio lower (.6), circ upper (1.6), circ lower (.8), perim conn (8), bkgmult lower (variable)
- First used PISA to examine the images with varying bkgmult lower (range: 2-9)
' | Bkgmult | Rolony Count | Notes |
NoDye_G680_Cy3 | 7 | 116 | |
NoDye_G680_Cy3 | 9 | 31 | |
NoDye_L2%_G680_FAM | 7 | 202 | picked up mostly nuclei |
NoDye_L2%_G680_FAM | 9 | 7 | |
dcProbe0-Cy3_G550 | 7 | 178 | |
dcProbe0-Cy3_G550 | 9 | 162 | |
dcProbe0-Cy3_G550 | 5 | 208 | |
dcProbe0-Cy3_G550 | 3 | 265 | decreasing bkgmult detects more "true" features without significantly counting artifacts |
dcProbe0-Cy3_G550 | 2 | 278 | doesn't detect anymore "true" features, just adds noise |
dcProbe0-Cy3_G680 | 7 | 544 | |
dcProbe0-Cy3_G680 | 9 | 489 | |
dcProbe0-FAM_L5%_G550 | 7 | 129 | |
dcProbe0-FAM_L5%_G550 | 3 | 209 | |
dcProbe0-FAM_L5%_G550 | 2 | 245 | still misses many while counting artifacts |
dcProbe0-FAM_G550 | 7 | 52 | |
dcProbe0-FAM_G550 | 5 | 68 | |
dcProbe0-FAM_G550 | 3 | 108 | even lowering it to 3 still won't count obvious features |
dcProbe0-FAM_G600 | 7 | 313 | |
dcProbe0-FAM_G600 | 5 | 364 | good |
dcProbe0-FAM_G680 | 7 | 519 | |
dcProbe0-FAM_G680 | 9 | 375 | not bad, pretty similar to G600 bkgmult5 |
Count Cells in Field of View
- NoDye_L2%_G680_FAM had nuclei with high intensity fluorescence relative to background, which was really obvious after processing in PISA so counted nuclei in that image
- 110 Nuclei/Cells counted
File:Agi26k0gap PGP1f InitialDetect NucleiCount110.png
Best Images and PISA Settings
' | Image Name | Image | Feature Count |
Cy3 Background | NoDye_G680_Cy3_Bkgmult9 | File:NoDye G680 Cy3 ch00 PISA7Overlay Bkgmult9.png | 31 |
FAM Background | NoDye_L2%_G680_FAM_Bkgmult9 | File:NoDye L2% G680 FAM ch00 PISA7Overlay Bkgmult9.png | 7 |
dcProbe0-Cy3 | dcProbe0Cy3_G550_Bkgmult3 | File:DcProbe0Cy3 G550 Cy3 ch00 PISA7Overlay Bkgmult3.png | 265 |
dcProbe0-FAM | dcProbe0FAM_G600_Bkgmult5 | File:DcProbe0FAM 600 FAM ch00 PISA7Overlay Bkgmult5.png | 364 |
Removing Background
- dcProbe0-Cy3
Ran PISA on NoDye_G680_Cy3_Bkgmult9 and dcProbe0Cy3_G550_Bkgmult3 Saved bw images in workspace as backgroundCy3 and dcProbe0Cy3 countOverlappedPoloniesv2(backgroundCy3,dcProbe0Cy3) ans = 0 comp_img1=zeros(xsize,ysize,3); comp_img1(:,:,1)=dcProbe0Cy3; comp_img1(:,:,2)=backgroundCy3; imwrite(comp_img1,'dcProbe0Cy3vsbackgroundCy3.png');
File:DcProbe0Cy3vsbackgroundCy3.png
- dcProbe0-FAM
Ran PISA on NoDye_L2%_G680_FAM_Bkgmult9 and dcProbe0FAM_G600_Bkgmult5 Saved bw images in workspace as backgroundFAM and dcProbe0FAM countOverlappedPoloniesv2(backgroundFAM,dcProbe0FAM) ans = 0 comp_img1=zeros(xsize,ysize,3); comp_img1(:,:,1)=dcProbe0Cy3; comp_img1(:,:,2)=backgroundCy3; imwrite(comp_img1,'dcProbe0Cy3vsbackgroundCy3.png');
File:DcProbe0FAMvsbackgroundFAM.png
- Conclusion: background is not a factor
Specificity of dcProbe0-Cy3 and dcProbe0-FAM
- Is there overlap between the two? (There shouldn't be, they should be exclusive)
countOverlappedPoloniesv2(dcProbe0Cy3,dcProbe0FAM) ans = 1
File:DcProbe0Cy3vsdcProbe0FAM.png
- The only overlap is in the bottom right corner and looks more likely due to nearby rolonies than non-specific binding of both dyes to some structure
- Red is dcProbe0-Cy3 and Green is dcProbe0-FAM
Overlay BrightField
File:DcProbe0Cy3vsdcProbe0FAM BF.png
Conclusion[edit]
Initial Imaging[edit]
- Background signal did not play a factor since none overlapped with detected features during dcProbe0-Cy3 and dcProbe-FAM
- We can say every feature we see is specific to 2nd rolony
- dcProbe0-Cy3 and dcProbe0-FAM showed essentially no overlap, which gives us confidence these are true signals
- Next time, we should be able to detect dcProbe1, dcProbe2, etc. and expect to see a fraction overlap with these
- dcProbe0-Cy3 had 265 rolonies (~2.4/cell)
- dcProbe0-FAM had 364 rolonies (~3.3/cell)
- dcProbe0-Cy3(3641 probes) + dcProbe0-FAM (4097 probes) = 7,738 probes/12,964 total probes in Agi26k0gap = 60%
- If these two dyes which can detect 60% of probes detected 629 2nd rolonies then we can approximate the total number of 2nd rolonies in this field of view = ~1050
- There are ~9.5 2nd rolonies/cell