| Overview and Guidelines |
 |
“What Every New Screener Should Know” Introduction to the Screening Process
Click here to view or download this document, which is an important
introduction to screening at the ICCB-Longwood facility.
Initiating a Screening Project at ICCB-Longwood
The ICCB-Longwood collaborates with academic investigators to perform
high-throughput screens of chemical libraries. Our compound collection
includes over 200,000 small molecules, including commercially
available compound libraries and unique natural product
libraries. We also offer genome-scale RNAi screening. Dharmacon SMARTpool siRNA libraries targeting the whole human genome
(21,176 genes) and the whole mouse genome (16,873 genes) are
available for screening at ICCB-L/NSRB. Data generated
from screens are entered into a screening results database. Access
to this non-public database is granted only to ICCB-Longwood investigators
and collaborators who have deposited data. For more information on small molecule data in the database,
please see the
ICCB-Longwood data sharing agreement.
The screening facility personnel assist efforts
by providing access to the compound and RNAi libraries, and by operating
and maintaining the screening robots. Collaborators perform their
own assays and provide their own supplies, including assay plates. A fee is charged for access for screening at ICCB-Longwood. Please see the small molecule application page and the RNAi application page for more details about the fee structures.
To allow us to prioritize new screening projects appropriately,
a formal application must be submitted in order to initiate a
screen at the ICCB-Longwood. Additional applications must be submitted each
time a new screen is proposed. We are currently reviewing applications
on a rolling basis. (Links to the applications appear in the previous paragraph or on the left hand navigator bar under Application Forms).
To avoid
situations in which two different groups are carrying out substantially
similar screens on the same small molecule collections at the same
time, we attempt not to initiate collaborations that compete
with existing ones. Whether a proposed screen will compete with
ongoing assays is decided by the ICCB-Longwood screening application Review
Committee in consultation with the investigators involved. Precedence
is generally given to the screen first proposed.
Radioactive assays
are currently not supported in the ICCB-Longwood screening facility,
and substances requiring a higher containment level than BSL2
are not permitted.
Please send questions about the application
process to screeniccb_apply"AT"hms.harvard.edu.
After
an application is accepted, the investigator is granted access
to the screening facility. Click here to link to the document "What
Every New Screener Should Know", which is requisite reading
before a screener's first
visit to the ICCB-Longwood Screening Facility.
Biosafety Requirements
The maximum biocontainment level for studies conducted in the screening facility
is BSL2.
Biosafety procedures for the screening facility are governed by Harvard
Medical School regulations, overseen by the Committee on Microbiological
Safety (COMS). The screening facility currently holds a biosafety approval
covering the items on our list of currently
approved organisms. If your screen
involves items not on this list, please contact us as soon as possible so that
we can file an amendment to add your reagents to the list.
To assist us in this process, if you are not a Harvard Medical School affiliated researcher, please provide a copy of your institutional
biosafety committee approval for all research procedures applicable to the
screen. Depending on the format of your institutional approval, you may have
to provide additional information describing your proposed work. For example,
the COMS application requests names and descriptions of strains, plasmids,
experimental design, and biosafety issues. Please review the HMS
COMS application form to see what
is required. Please be aware that COMS approval may take a month or more, so
it is important to file amendments as early as possible.
Your BSL2, hazardous
waste, and bloodborne pathogen training should also be completed and up-to-date.
Annual refresher courses for Harvard personnel are available online at University
Operations Systems.
For screeners not holding current Harvard IDs, access to the screening facility
is contingent upon signing a statement that your employer is legally responsible
for providing all necessary BSL2, hazardous waste, and bloodborne pathogen
training and protective equipment. Please check with your employer regarding
these issues. For practical purposes, however, site-specific training and
protective equipment will be provided by the screening facility as necessary. Non-Harvard screeners must sign a biosafety training statement and give it to the screening facility staff. Please download the PDF by clicking here: biosafety training statement.
Designing a High
Throughput Assay
Assay Volume and Plate Type
For a high-throughput screen,
a biochemical or cell-based assay must be adapted to a microtiter
plate format. Assays are most commonly performed in 384-well assay
plates; in certain cases, they may also be performed in 96-well. Assay volumes in 384-well plates range from
5 µl (in low-volume plates)
to 100 µl (in standard plates). At the low end of this range,
inaccuracies in small-volume pipetting can cause signal variation,
while there is a risk of spillage and cross-contamination at the
high end. We therefore recommend that investigators use 30 µl
assays in standard plates if possible.
Different types of plates
are recommended, depending on the assay detection method used.
White plates are recommended for luminescence assays, black plates
for fluorescence assays, clear-bottom plates for automated microscopy,
and clear polystyrene plates for photometric assays. Please see
the Screening Supplies section for recommendations on
plates.
Replicates and Controls
Most high-throughput assays show a high
amount of inherent variability and error. Therefore, it is strongly
recommended that all assays be performed in duplicate if possible.
The best method is simply to duplicate the entire screen in a
new set of assay plates. This is far more reliable than re-analyzing
or re-reading the same assay plates twice. Using duplicate data
points can reduce false positive rates by up to one-half.
Control
readings are essential to a well-designed assay, and every assay
should use as many controls as possible. In general, there are
two types of controls: plate-based controls and assay-wide controls.
Plate-based controls are controls that are placed on each individual
assay plate. These are essential for identifying plate-to-plate
variability and establishing assay background levels. Assays
that are prone to plate-wise variability (such as luciferase readouts
that decay over time) should primarily use plate-based controls
and normalization. Stock library plates are formatted
with empty wells to allow for screen-specific plate-based controls. It is good
practice to use all available wells, with the researcher deciding
on the appropriate distribution of positive or negative controls.
Assay-wide
controls are separate plates containing only control wells and
no screening compounds. These are particularly useful for determining
the background levels of an assay and should be used to help
determine whether an assay has sufficient signal to be reliably
detected.
Signal/background ratio (S/B), and well-to-well variability
(CV) are important issues to consider. As assay variability increases,
the S/B ratio must increase for the screen to be successful.
We recommend using a positive control condition to determine the
S/B ratio. To determine S/B, fill a plate with reagents using the
same equipment to be used for the screen. Add several dilutions
of the positive control to several wells, and determine whether
the positive control can be reproducibly detected above the well-to-well
variation. These data will provide an indication of the false-positive
and false-negative rate of your assay.
Quantitative Assay Evaluation and Optimization
Click here to download the Quantitative Assay Evaluation and Optimization information as a PDF file.
The Z´-factor calculation is useful during piloting for quality assessment of assay conditions (Zhang et. al. 1999, see below for full reference). An assay can be considered validated for high-throughput screening after 3 independent experiments (for example, each experiment set up separately starting from scratch or carried out on different days) have been shown to result in reproducible and suitable Z´-factor values. Each experiment should be performed on at least 1 full 384 well plate where ½ of the wells contain positive controls and ½ of the wells contain negative controls. This will produce a statistically significant data set for evaluation.
To quantitatively rank assay conditions, perform control experiments and calculate Z´ from the data collected:
SD + = positive control standard deviation
SD - = negative control standard deviation
Ave + = positive control average
Ave - = negative control average
Note – Microsoft Excel has a STDEV function that works well for this calculation
After calculating Z´ the following table is useful for evaluating the potential performance you might expect using this assay to screen at the ICCB-L/NSRB screening facility. If optimization is needed, different assay conditions should be compared and ranked by their Z´-factor values until suitable conditions are found.
High-throughput Screening Assay Fitness Table
For small molecule screening assays, the below Z' values indicate assay fitness:
1 > Z´ > 0.9 An excellent assay
0.9 > Z´ > 0.7 A good assay
0.7 > Z´ > 0.5 Hit selection will benefit significantly from any improvement
0.5 = Z´ The absolute minimum recommend for high throughput screening
This table may differ slightly from published recommendations. However it is based on the general experience of small molecule screeners in our facility. We commonly observe that screening results rarely achieve the high quality levels seen during piloting using controls.
For RNAi screening assays, the experience of ICCBL screeners has been that Z' factor values under the most optimized conditions are often less than 0.5. These RNAi screening assays have nonetheless been productive. However RNAi screeners should consider the possibility of carrying out their screen in triplicate to increase robustness.
Please contact us if you require more information. Click here to download the Quantitative Assay Evaluation and Optimization information as a PDF file.
Z'-factor Reference: Ji-Hu Zhang, Thomas D. Y. Chung and Kevin R. Oldenburg (1999). A simple statistical parameter for use in evaluation and validation of high throughput screening assays. J. Biomol. Screen 4:67-73.
Performing a Screen
Once a satisfactory assay is developed in a 384-well assay plate
format, contact the ICCB-Longwood screening facility staff (iccb_screen"AT"hms.harvard.edu or 617-432-5815) to schedule your screen. Please contact the screening
personnel well in advance of the date you would like to start your
screen, as the screening schedule may be filled several weeks in
advance.
Screening facility personnel will provide screeners with
training in the use of automated liquid handlers, plate readers,
and screening microscopes. Only screening facility personnel
are permitted to handle library stock plates, and thus they perform
all transfers from library plates into assay plates.
Screening facility personnel are not responsible for conducting
any other parts of a screen.
After establishing an assay protocol,
you should perform a pilot screen of 2,000 to 3,000 wells.
Additional information about libraries available for screening
is available at Compound
Libraries and RNAi libraries.
Note that we provide an up-to-date list
of recommendations for the order in which to screen our many small molecule libraries. If the
results of your pilot screen are satisfactory, you may continue
to screen additional library plates. At this point, you must submit
an assay protocol for inclusion in our screening database.
When
performing a screen, it is preferable to process as many plates
as possible during a single visit to the screening room. We strongly
suggest screening all wells in duplicate to increase the
significance of data points generated. The typical daily throughput
for an individual small molecule screening project is approximately 20 plates
in duplicate (40 plates total, or roughly 14,000 wells) for screens
involving mammalian cells, and somewhat higher for microbial screens.
An average ICCB-Longwood screening project assays a total of 50,000 to 100,000
compounds for biological activity.
ICCB-Longwood provides screening libraries
and access to equipment but does not provide supplies for screeners.
Screeners are responsible for purchasing and shipping all applicable
supplies to ICCB-Longwood for use in their screens.
Data Handling
Raw screening data should be saved directly to your ICCB-Longwood server
account. Detailed instructions for data formatting and deposition
into our screening results database may be viewed here (PDF). Once data are entered into the database, instruction
will be provided as to how to view your data and compare your results
to those of other screens.
Your data must be entered into the database
in a timely manner after completing your primary screen. Selection
of compounds/siRNAs for secondary assays ("cherry picking")
is permitted only after deposition of initial data. Please see
below for additional information on cherry picking.
While it is
the responsibility of individual researchers to analyze their
screen results, ICCB-Longwood staff can provide advice during this process.
Some information may be viewed under the Data Analysis section:
General Considerations and
Interpretation
of Hits.
Small Molecule Compounds for Secondary Screens
After completing a primary screen of a library, you should have
many screening "positives" to follow-up in secondary
screens. Subject to compound availability, we will provide you
with 1 µl of each selected screening positive compound (~5 µg).
These are called “cherry picks”. The maximum number
of cherry picks allowed is equal to 0.3% of the number of compounds
initially screened. For example, if the ChemBridge DiverSet E library
is the initial screen completed (16,320 compounds), a maximum of
50 cherry picks can be requested. With adequate justification,
exceptions to this policy can be made.
Screeners with a "hit" rate
that is higher than 0.3% will often choose their cherry picks by
examining the structures of their screening positives and comparing
these data with others' screening data. This requires data entry
into the screening database and also time spent by the researcher
mining the database. This step is usually important for the success
of the project.
Cherry-pick requests should be sent to David Wrobel
and are generally supplied within two weeks. Please view Cherry
Pick Guidelines (PDF) for instructions about submitting a cherry pick request.
If you
require more than 1 µl of compound, you must order
it directly from the original source. Generally, 1-5 mg of a compound
can be purchased for less than $50 from the manufacturer for commercial
libraries. Commercial sources are often able to
resupply a majority of their compounds for the first year after
a library is purchased. The resupply rate falls for older libraries
(e.g.. as of April 2002, the ChemBridge DiverSet E collection had
a resupply rate of only about 75%). Please see the descriptions
of the individual Compound
Libraries for information on ordering.
For compounds no longer available
from the original vendor, other supply sources may be found using
SciFinder or ChemNavigator. See the Interpretation
of Hits section. Alternatively, custom synthesis can be arranged
to obtain additional compound. Rates for custom synthesis vary
considerably and must be negotiated directly with suppliers.
Other Questions
Please visit the FAQ section for answers to commonly asked screening questions, and click here to see information on how to build an academic high-throughput screening facility. If you have additional questions regarding the screening process or policies, contact the screening facility staff at iccb_screen"AT"hms.harvard.edu or Caroline Shamu.
Click here (PDF) to view “What Every New Screener Should Know”, suggested reading before a screener's first visit to ICCB-Longwood.
|
top
of page