Biology

Designing a Screening Campaign:
Goals, Approach and Plan

A critical question for developing a screening campaign is, “What are your goals and how do you define success?” You may be looking for a research tool to characterize a molecule of interest, identifying a novel target or modeling a precise three dimensional network of interactions at a key protein-protein interface. The Center’s assay development and optimization services encompass many aspects of cell-, protein- and organism-based models, including assessments of enzymatic activity, protein interactions, cell-based readouts and high content cell imaging. We’ll work with you to understand your goals to customize a project that meets your needs.

We work with our partners to combine their deep knowledge of the field with the broad screening experiences of Center. A YCMD staff member is paired with a team member from our partner’s group to develop an overall strategy for success, including an assessment of resources required, screening approaches, data management and interpretation. Together, the team generates novel materials and information to support grant applications and/or to support potential downstream partnering opportunities.

During our initial conversations, a work plan is developed to identify source material, staffing, unique reagents and a biosafety plan. We implement these to evaluate feasibility for high throughput screening and then work to optimize the robustness that is required for success.

Screening

Assay Development: The Center actively seeks to learn more about your research and screening goals and to develop a robust and practical strategy and to optimize the approach for high-throughput (and high content, as needed) screening. We work with you to design and optimize your assay. We use statistical measures of assay robustness, namely Signal to Background Ratio, Z’ and Coefficient of Variations for control populations, to assess screen readiness.

The Center supplements a large component of the screening process, underwriting many of the costs required for Center staff and equipment. Most the equipment and experiences already exist within the Center.

Primary Screening: Primary screening typically generates single point data using one concentration for each compound and triplicate data points per siRNA. We normalize screening relative to the control populations on each plate, and provide plate-based summary statistics for an evaluation of screen performance. Hit selection strategies to fit your project goals are implemented. Information from external databases, such as The Gene Ontology, may be merged with screen data to facilitate hit identification in siRNA screens.

Investigators are responsible for procuring reagents to perform their screen whereas the Center is responsible for standard laboratory consumables (pipet tips, assay plates). The price for setting, running and analyzing a primary screen typically reflects a combination of a fee per assay plate plus an hourly fee when Center staff are needed. These fees are largely subsidized by the Center.

For a full description of our screening collections visit our Screening Collections page.

Hit Picking and Dose-Response: Hit picking is performed after a primary screen to retest active compounds and to rule out compounds which exhibit assay interference or toxicity, or to validate siRNA duplexes for their effects. Dose Response studies, using serial dilutions of each compound, can be performed within a plate, or across a series of plates in cases where many compounds are being tested. For intraplate dose response, concentrations often range from 50 uM to 0.3 nM to generate 12-point curves. Thus, the dose-responsiveness of a dozen compounds (and proper controls) can be evaluated in each 384-well plate. Hit selection strategies to fit your project goals are implemented. We can fit the data to appropriate dose-response models and extract fit parameters for lead prioritization.

The price for setting up, running and analyzing the results is a combination of a fee per assay plate and a Center use hourly fee, which are subsidized by the Center.

Image Analysis: Features of complex cellular phenotypes captured in microscopy images can be identified and quantified using high throughput methods. In most cases, nuclei are identified first, and other cellular features dependent upon that nuclei assignment are delineated. Size,shape, signal intensity and number, are just a few of the features which can be analyzed from fluorescent or transmitted light multi-channel images. We use statistical measures on analyses derived from replicate images of control treatments to assess algorithm performance.

The price for developing an algorithm, processing images and analyzing the results is based on a Center use hourly fee, which is subsidized by the Center.

Center Equipment Use

Capital equipment in the Center can be made available to qualified users of the Yale community on a case-by-case basis. The equipment includes different instrumentation for liquid handling or high throughput readouts. Potential applications could include re-arraying (e.g., from 96 to 384 well plates); replica plating; pin-tools to transfer nanoliters of liquid reagents; plate washing and preparation of samples for microscopy. High content screening (HCS) capabilities include low and high throughput platforms. Please contact us to discuss your needs. Work is charged an hourly rate.

Screen Follow-up and Compound Selection

High throughput or high content data collection requires a series of steps to triage for the most promising leads.

  1. Establish activity thresholds. Active molecules are those with activity values above an objective, measurable threshold. Once this threshold is established, these candidates are termed “screen actives.” By combining the effective concentrations and magnitude of overall activity, the team can identify promising chemical matter.
  2. Filter potential hit list. Screening libraries invariably include compounds, which interfere with certain assays in a non-specific manner. Certain compounds are flagged as “frequent hitters” based on reactive groups or behavior in assays and thus are de-prioritized for further investigation.
  3. Confirm Activity and Exclude Detection Artifacts. To ensure candidates are reproducibly active, results are reproduced in single point or dose-response studies. In parallel, controlled experiments exclude candidates that interfere assay parameters.
  4. Observe dose-dependence. Dose response curves are examined for unusual behavior. This is termed “Hit Picking” phase and allows the investigator to advance compounds that consistently function in a dose-dependent manner.
  5. Computational analysis of validated hits can sort candidates into “clusters” by relating chemical structure with screen-based activity. Cluster analysis can validate a biochemical target and extract structure-activity relationships (SAR). The resulting computational models can provide utility for “virtual screening” in silico, and thereby predict novel molecules for further study. Likewise, cluster analysis can identify false-positives and singletons (which do not belong to any cluster) for special consideration. In addition, calculation of ligand efficiency may productive binding interactions of particularly small molecules, which might have been overlooked based on their small size.
  6. Secondary assays or whole-cell screens further validate activity and control for artifacts that might have arisen due to the primary screen. Behavior in whole-cell screens may also reveal cell permeability issues that could be problematic for certain applications.
  7. Chemical confirmation of structure allows one to identify changes. Compounds may oxidize, hydrolyze and rearrange on storage in solution state. This is typically accomplished with HPLC and mass spectroscopy (LC/MS).
  8. Neat samples procured. To verify that the suspected molecule is responsible for the desired activity, commercially available samples are ordered, re-synthesis begun or de-novo synthesis planned. In some cases, these compounds may be generated within the Center.
  9. Virtual screens conducted. Cluster and SAR analysis of validated hits is used to construct a chemical profile of your hits. This profile serves as a fingerprint to match with other databases of compounds for structures likely to be active. These virtual hit molecules are procured and tested for activity. This virtual screen and real-screen follow-up may be repeated and thereby refined.
  10. Determine chemical lead series. A chemical “lead” is a term used to define a compound or series of compounds defined by a recognizable common scaffold, which display certain activities against a biological target. The ideal chemical lead series will have a set of physical, pharmacokinetic and toxicological properties appropriate for the intended purpose (example: plant or animal dosing). Factors affecting the suitability of a lead include solubility, cell permeability, in-vivo stability, synthetic accessibility. The chemistry optimization plan based on the lead would be customized depending on investigator’s objective. IP novelty may be determined at this point by literature searches.

Please contact Janie Merkel (janie.merkel@yale.edu) for the current fee structure applicable to your project.