The main planned activities are the following, organized in short and long term: 

 Short term (1 year): 

• Collection of IHC validation virtual slides already produced in the participants’ core labs o Creation of a middle-sized curated online dataset hosted in OMERO or GCP that can be accessed by the scientific community that includes a manual evaluation of the antibody performance 
• Creation of a troubleshooting dataset regarding experience with different instruments and kits, advantages and disadvantages, uses and limits, etc. 
• Contact the reference people of HubMap and other IHC-focused groups to join forces, exchange visions, planned work and avoiding re-inventing the wheel 

Long term (more than 1 year): 

• Expansion of the online dataset to a larger size (including transition to a different hosting) o Bioinformatic and deep learning analysis of staining pattern on the collected dataset images (to be developed) 
• Collection of human and other species tissues and cells validation samples for IHC/mIF and for other validation purposes (mass spectrometry-based validation, overexpression, KO, etc.) 
• Experimental studies for validating specific groups of antibodies deploying different approaches like overexpression, KO, MS-based proteomics, etc. (use-focused validation, section to be better defined) 

For the planned activities of above we have the following resources already available: 

• An OMERO dataset already owned and maintained at NYU Experimental Pathology core lab (Valeria Mezzano) that can be used for hosting the first version of the annotated dataset 
• A Google Cloud Platform available at Weill Cornell Medicine, department of Pathology, MISI core lab, that could be used for hosting a larger version of this dataset and for performing deep learning-based analysis and other bioinformatic analysis 
• A dataset of 18 human cell lines for which is available a FFPE cell microarray, mass spectrometry-based quantification of the proteome and RNA-sequencing data (Fabio Socciarelli) 
• Instrument capability available from the MISI core of Weill Cornell Medicine (Bond RX stainer, Vectra Polaris  slide scanner, COMET Lunaphore Instrument) and from other institutions histology imaging core labs  participating to the project (NYU, MSKCC, Vanderbilt University Medical Center, Fred Hutch Cancer Center,  University of Iowa, University of Michigan, University of Colorado, University of Oklahoma, University of  Virginia, Van Andel Research Institute, Janelia, Mayo Clinic, MIT, The Jackson Laboratory and other  institutions) 
• Collection of validation images already produced and scanned in the institutions participating to the study (see point above for the main institutions, the estimation of images available is already ongoing) • Access to the FFPE tissue archive of pathology of the New York Presbyterian Hospital for Quality Control/Quality Assurance purposes (Fabio Socciarelli, IRB exemption already obtained for the specific purpose, amount of samples in the order of hundreds of thousands) 
• Availability of the archive of mouse tissues and other species at University of Michigan (Ingrid Bergin) 

While the above resources are valuable and sufficient for the short-term aim of producing a medium-sized annotated dataset for IHC/mIF, they will not be enough for building a large size dataset and for the creation of new wet lab data for validation. The main perceived constraint is the lack of funding that we will need to tackle (application to grants and other sources of funding) and the help of other expertise that are missing in this group for the long-term goals (AI expertise, cell-based validation, etc.)