Optimized Experimental and Analytical Tools for ...
Transcript of Optimized Experimental and Analytical Tools for ...
![Page 1: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/1.jpg)
Optimized Experimental and
Analytical Tools for
Reproducible Drug-Response Studies
Caitlin Mills & Luca Gerosa
Department of Systems Biology
& Laboratory of Systems Pharmacology
Harvard Medical School
![Page 2: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/2.jpg)
Optimized Experimental and
Analytical Tools for
Reproducible Drug-Response Studies 1 2
4
3
Caitlin Mills & Luca Gerosa
Department of Systems Biology
& Laboratory of Systems Pharmacology
Harvard Medical School
![Page 3: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/3.jpg)
In-vitro drug-response studies in cancer are often based on relative cell number quantifications
Drug concentration (M)
Rela
tive c
ell
num
ber
(Tre
ate
d/C
ontr
ol)
Cell line X, 72 hrs
![Page 4: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/4.jpg)
Drug concentration (M)
Re
lative
ce
ll n
um
be
r
AUC
Drug-response in cancer research and conventional metrics based on relative cell number
![Page 5: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/5.jpg)
Irreproducible pharmacogenomics and other
drug-dose response based studies
1. CCLE & GDC, Nature, Dec 2015
2. Haverty et al., Nature, May 2016
3. Bouhaddou et al. Nature, Dec 2016
4. Mpindi et al., Nature, Dec 2016
5. Safikhani et al., Nature, Dec 2016
6. Geeleher et al., Nature, Dec 2016
![Page 6: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/6.jpg)
Irreproducible pharmacogenomics due to
irreproducibile IC50 and other metrics
Inconsistency in large pharmacogenomics studies, Haibe-Kains et al, Nature, 2013
Drug-dose response correlation between Cancer Genome Project(CGP) and
Cancer Cell Line Encyclopedia(CCLE):
![Page 7: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/7.jpg)
Rethinking drug-dose response metrics
Marc Hafner
Drug dose=0.1 uM Drug dose=3 uM
Drug concentration (M)
Rela
tive c
ell
num
ber
![Page 8: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/8.jpg)
Normalized growth rate inhibition (GR) value
k(c) is the treated growth rate
k(0) is the control growth rate
![Page 9: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/9.jpg)
GR values rely on three measures of cell count
x(c) is the treated cell count
xctrl is the control cell count
x0 is the cell count at the
time of treatment
![Page 10: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/10.jpg)
GR values are independent of the division rate
and directly relate to the phenotype
GR metrics:
GR50, GRmax and GRAOC substitute for IC50, Emax and AUC
![Page 11: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/11.jpg)
GR metric allows for an intuitive assessment of
phenotypic effects across cell lines and drugs
34 breast cancer cell lines
![Page 12: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/12.jpg)
GR metric allows for an intuitive assessment of
phenotypic effects across cell lines and drugs
![Page 13: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/13.jpg)
Cell seeding affects division time which biases
traditional sensitivity metrics
Seeding density affects the number of divisions.
IC50 and Emax are correlated with density.
Spearman’s correlation with seeding number (11 drugs)
![Page 14: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/14.jpg)
Genetic alterations affect division time which
biases traditional sensitivity metrics
Etoposide sensitivity in HME RPE-1 cells with inducible
BRAFV600E expression.
Thanks to Jia-Yun
Chen for the cell line
![Page 15: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/15.jpg)
Genetic alterations affect division time which
biases traditional sensitivity metrics
![Page 16: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/16.jpg)
Time-dependent GR metrics
For evaluating GR50(t) and GRmax(t) and quantifying
adaptive response or late drug action.
![Page 17: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/17.jpg)
Time-dependent GR can reveals dynamic
changes in drug-response effects
Time-dependent GR
Time (days) Time (days)
DMSO
RAFi
RAFi+
MEKi RAFi+
MEKi
RAFi
Cell number
BRAFV600E melanoma cell line A375
GR (72 hrs)
RAFi RAFi+
MEKi
Cytotoxic
Cytostatic Cytostatic
![Page 18: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/18.jpg)
Decoupling cytostatic and cytotoxic drug-
responses by GR metrics
Viable cells
x(c,t)
x(0,t)
x(0,0)
Dead cells
d(0,0) d(0,t)
d(c,t)
Normalized growth rate:
Normalized death rate:
Decoupled GR metric:
![Page 19: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/19.jpg)
Similar drug responses can be due to different
combinations of cell growth and death
(Hsp90 inhibitor)
=
![Page 20: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/20.jpg)
Re-Analysis using GR metric of Genentech Cell Line Screening Initiative
(409 cell lines and 16 drugs):
False
positives
False
negatives
GR metrics correct growth rate confounders in pharmacogenomics and reveal true associations
![Page 21: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/21.jpg)
False negative example: PTEN mutate ARE insensitive to Lapatinib
False positive example: ΔCDC73 are NOT sensitive Docetaxel
GR metrics correct growth rate confounders in pharmacogenomics and reveal true associations
![Page 22: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/22.jpg)
GR metrics improve reproducibility between
pharmacogenomics studies
Comparison of drug-dose response for 9 drugs and ~100 cell lines in the
Genentech Cell Line (gCSI) and Cancer Therapeutic Response Portal (CTRP) studies:
![Page 23: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/23.jpg)
Conclusions on GR metrics as analytical tools
for reproducible drug-dose responses
GR metrics...
• ... eliminate confounders that act by cell division bias (cell
seeding, genetic background, etc...)
• ... can be extended to quantify time-dependent drug-response
and to decouple cytostatic and cytotoxic effects
• .. improve reproducibility in studies that rely on measuring
growth inhibition, such as in pharmacogenomics
![Page 24: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/24.jpg)
Optimized Experimental and
Analytical Tools for
Reproducible Drug-Response Studies 1 2
4
3
Assay development
Example from our work
5
6
![Page 25: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/25.jpg)
Experimental pipeline
1. Designing experiment
2. Running experiment
3. Processing data files
4. Evaluating sensitivity
metrics
Hafner*, Niepel*, Subramanian*, Sorger
Curr Protoc Chem Biol, 2018
![Page 26: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/26.jpg)
To consider before you start: cell lines
• How many cell lines do I want to test?
• Are they amenable to imaging?
• Are they adherent?
• Do they grow in a monolayer?
• How densely should they be seeded?
cell lines
![Page 27: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/27.jpg)
To consider before you start: drugs
• How many drugs do I want to collect dose response
data for?
• Are they DMSO soluble?
• How many dose points do I need?
• What’s an appropriate dose range?
• How many time points do I want to test?
• How long should the assay run?
• What are the expected effects of drug treatment?
drugs
time
doses
![Page 28: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/28.jpg)
To consider before you start
96 or 384 well plates
![Page 29: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/29.jpg)
Do I need to use the GR approach?
It depends on how you answered the previous
questions.
Relative cell counts are valid when the untreated
controls do not change:
• Phenotype is not related to cell growth
• Untreated cells do not grow
• Short assays during which growth is negligible
![Page 30: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/30.jpg)
Experimental design
![Page 31: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/31.jpg)
Design scripts available at github.com/datarail/datarail
![Page 32: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/32.jpg)
![Page 33: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/33.jpg)
Metadata table
Treatment layout
Treatment file
![Page 34: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/34.jpg)
Set up library plates for pin transfer for large
scale experiments
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Randomized Library Plates Manual layout of drugs on source plates
Use controls to ‘barcode’ library plates.
![Page 35: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/35.jpg)
Use other automation for pilot, follow-up and
smaller experiments
![Page 36: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/36.jpg)
Design steps to improve reproducibility
• Randomization of the treatments across multiple
technical replicates
• Standardize nomenclature, barcode plates
• Control for plate bias (across day 0 plate; positive &
negative controls across treatment plates)
• Robotic treatments with the D300 or pin transfer
• Exclude edge wells whenever possible
![Page 37: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/37.jpg)
Randomization can mitigate edge effects
![Page 38: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/38.jpg)
Running the experiment
![Page 39: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/39.jpg)
Experimental design is complete. Now what?
• Grow (happy) cells
• Seed cells at appropriate densities in multi-well plates
• Deliver drugs to multi-well plates
• Stain and fix cells
• Image cells
• Extract quantitative data from images
![Page 40: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/40.jpg)
Cell seeding
• Seed plates at an appropriate density from parent
plates in log-phase growth
• Use automation if possible
• Barcode plates to keep track of them
![Page 41: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/41.jpg)
Cell seeding density influences growth rate...
![Page 42: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/42.jpg)
...which influences the dose response
![Page 43: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/43.jpg)
Drug delivery via pin transfer
• For simultaneous delivery of many drugs
• For large scale experiments (many cell lines,
conditions)
• Facilitates reproducibility
![Page 44: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/44.jpg)
Drug delivery via digital drug dispenser
• For accurate delivery of a few drugs
• Pilot experiments- to identify appropriate doses
• Follow-up experiments, ‘hit’ validation
• Drugs that cannot be prepared in DMSO
![Page 45: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/45.jpg)
No automation? Use serial dilutions
400 L
media
400 L
media
400 L
media
400 L
media
400 L
media
1200 L
1000 nM
drug in media
Transfer
700 L
Transfer
700 L
Transfer
700 L
Transfer
700 L
Transfer
700 L
![Page 46: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/46.jpg)
and multichannel pipettes
636 nM
drug
405 nM
drug
258 nM
drug
164 nM
drug
104 nM
drug
1000 nM
drug in media
![Page 47: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/47.jpg)
Dye-drop assay reagents
• Minimally-disruptive, reagent-sparing cell staining
and fixation protocol
![Page 48: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/48.jpg)
Dye-drop assay protocol
• Stain: Hoechst + LDR in 10% optiprep in PBS
• Fix: 4% formaldehyde in 20% optiprep in PBS
![Page 49: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/49.jpg)
Plate washer
• Uniform and controlled aspiration and liquid
dispensing
• Is repeat washing really that bad?
![Page 50: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/50.jpg)
Repeat washing can result in cell loss…
No wash PBS wash x 1 PBS wash x 2
![Page 51: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/51.jpg)
…that can bias your results
![Page 52: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/52.jpg)
Image acquisition
• Operetta microscope with plate hotel, barcode
reader & robot
– Automated data collection for 40+ plates
![Page 53: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/53.jpg)
Image acquisition
Imaging 6 fields of view @ 10x captures almost the entire well
![Page 54: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/54.jpg)
Image acquisition
![Page 55: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/55.jpg)
Image analysis 1. Segment nuclei
2. Measure LDR signal
3. Classify live/dead cells
![Page 56: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/56.jpg)
Can I just count live cells?
![Page 57: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/57.jpg)
Strengths and limitations of the dye-drop assay
• Imaging based
– Best suited for adherent cells that grow in monolayer
culture
• Image analysis can be time consuming
• Can go back and visually inspect imaging data
• Potential for multiplexing, immunofluorescence
• Reagent sparing
• Distinction between cytotoxic and cytostatic effects
• Fate of live cells unknown
![Page 58: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/58.jpg)
Other common dose response assays
• CellTiter-Glo etc. – Simple, no wash protocol – Luminescence read-out, simple analysis, rapid results – Treatment-induced changes in metabolic activity of cells can
skew results
• Measurement of confluency – Inaccurate – Treatment-induced changes in morphology can skew results
![Page 59: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/59.jpg)
Example of artefact with a CDK4/6 inhibitor
Rela
tive v
iabili
ty
1 µM Palbociclib
DMSO
![Page 60: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/60.jpg)
Take away messages
• Include a t=0 plate
• Optimize conditions, be consistent
– Seeding density per cell line
– Dose range per drug
– Duration of the assay
• Automate as much as possible
![Page 61: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/61.jpg)
Data processing
![Page 62: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/62.jpg)
Data analysis
![Page 63: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/63.jpg)
Raw quantified image data
Metadata from design
![Page 64: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/64.jpg)
Normalized count table
GR values
Summary GR metrics
Summary plots
![Page 65: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/65.jpg)
Analysis output files
Normalized count table
GR values
GR metrics
![Page 66: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/66.jpg)
Online GR tool: www.grcalculator.org
![Page 67: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/67.jpg)
![Page 68: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/68.jpg)
![Page 69: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/69.jpg)
Compound A Compound B Compound C
Compound E Compound D
![Page 70: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/70.jpg)
Optimized Experimental and
Analytical Tools for
Reproducible Drug-Response Studies 1 2
4
3
Assay development
Example from our work
5
6
![Page 71: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/71.jpg)
Profiling the responses of triple negative breast
cancer models to kinase inhibitors
• Why study kinase inhibitors in TNBC?
– Unmet clinical need
– Patients have a poor prognosis, and no targeted
therapy options
• GR metrics were used to enable comparisons
across cell lines
![Page 72: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/72.jpg)
Selection of cell lines and drug treatments
20 TNBC
6 HR+
4 Her2amp
2 NM
4 from PDX
X
24 kinase
inhibitors
4 misc
inhibitors
3 chemo
4 DNA
damage
![Page 73: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/73.jpg)
Data collection workflow
72 hrs
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
A
B 1 8
C 2 9
D 3 29 10 30
E 4 11
F 5 12
G 6 13
H 7 14
I 15 22
J 16 23
K 17 24
L 18 25
M 19 26
N 20 27
O 21 28
P
A B
C D
36 hrs
![Page 74: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/74.jpg)
Dose response results for one cell line
![Page 75: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/75.jpg)
Dose response results for all cell lines
![Page 76: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/76.jpg)
Diversity in response profiles…
![Page 77: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/77.jpg)
results from cytostatic and cytotoxic effects and…
![Page 78: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/78.jpg)
…occurs in both potency and efficacy
across cell lines and drugs
We aim to understand the biology underlying these differences with the goal of being
able to predict the response of a cell line to a perturbation.
![Page 79: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/79.jpg)
Can we learn more about the live cells?
Deep dye-drop assay development
![Page 80: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/80.jpg)
Image acquisition is more time consuming
Hoechst LDR
p-H3 EdU
![Page 81: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/81.jpg)
Image acquisition
Merge
Hoechst
LDR
p-H3
EdU
Cell count
DNA content
Dead cells
S phase cells
M phase cells
![Page 82: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/82.jpg)
Effects of drug treatment on cell cycle
Untreated Treated
![Page 83: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/83.jpg)
Strengths and limitations
• Deeper phenotype
• Characterization of surviving cells, cell cycle effects
• Flexibility in antibody selection
• Increased cost
• Longer image acquisition time
• Work with object level data required
![Page 84: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/84.jpg)
Conclusions
• Planning, and optimization promote reproducibility
• Automate as much as possible, know how it works
• Script the experimental design and analysis
• Use appropriate metrics for your experiment
![Page 85: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/85.jpg)
Acknowledgements
Sorger Lab/LSP:
• Marc Hafner
• Mirra Chung
• Jeremy Muhlich
• Mario Niepel
• Kartik Subramanian
• Nick Clarke
• Liz Williams
• Peter Sorger
Funding:
• NIH LINCS grant
U54-HL127365
ICCB-L:
• Stuart Rudnicki
• Caroline Shamu
• Richard Siu
• Jen Smith
• Rachel Warden
![Page 86: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/86.jpg)
Contact us with questions
• grcalculator.org
• github.com/datarail/datarail
![Page 87: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/87.jpg)
Useful references
• Hafner*, Niepel* et al. Nat Methods, 2016, 13(6):521-7
• Hafner, et al., Nat Biotech, 2017, 35(6):500-2
• Hafner*, Niepel*, Subramanian* et al., Curr Protoc
Chem Biol, 2018, 9(2):96-116
• Niepel*, Hafner* et al., Curr Protoc Chem Biol, 2018,
9(2):55-74
![Page 88: Optimized Experimental and Analytical Tools for ...](https://reader033.fdocuments.in/reader033/viewer/2022042211/62596f0b3eed3114b6718e71/html5/thumbnails/88.jpg)
Homework assignments
iccb.med.harvard.edu Assay Automation and Quantitation 2018
4. HTS Group 1 – visualization and comparison of two cell-based small molecule screens
that are looking for compounds that are selectively positive in one of the two cell lines tested
5. HTS Group 2 – visualization and comparison of two molecule screens, one is biochemical
and looking for compounds that disrupt protein-protein binding and the other is trying to
identify compounds that are selectively positive in one of the two cell lines tested
1. Drug Response Problem Set 1 – data processing and quality control; complete a typical
workflow for processing and analyzing HT drug response data
2. Drug Response Problem Set 2 – data reproducibility; compare the results obtained by
different labs that screened the same set of compounds in the same cell line
3. Drug Response Problem Set 3 – variability in drug response; evaluate the distribution of
the calculated drug-sensitivity metrics when ~70 cell lines are treated with ~100 small
molecules