> For the complete documentation index, see [llms.txt](https://docs.owkin.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.owkin.com/getting-started/why-k-pro.md). # Why K Pro? How K Pro differs from other AI-for-target-prioritization platforms, and the outcomes it is designed to improve. This page explains how K Pro differentiates from other platforms in the AI-for-target-prioritization space. For an overview of what K Pro does, see [What is K Pro?](/getting-started/home.md). ### How K Pro differentiates 1. **Multimodal patient data at depth.** Most platforms sit on top of aggregated public databases (Open Targets, GWAS Catalog, ChEMBL). K Pro is built on Owkin's data network of 164 academic medical institutions, with MOSAIC as the flagship cohort in oncology: one of the largest spatially-resolved multimodal oncology datasets, combining bulk RNA-seq, single-cell RNA-seq, spatial transcriptomics, whole-exome sequencing, H\&E and IHC pathology imaging, and longitudinal clinical follow-up. For target prioritization, this enables spatial co-localization of targets with TLS structures, malignant-cell specificity at single-cell resolution, intra-tumor heterogeneity quantification. 2. **Specialized AI models, wrapped in expert skills.** K Pro orchestrates calls to specialized models like HistoPLUS (Owkin's computational pathology foundation model, Adjadj et al., 2025), which detects 13 cell types including under-studied immune populations like neutrophils and eosinophils, with 14% better F1 classification than prior SOTA at 5× fewer parameters. It can also call Owkin Zero, our proprietary biological reasoning model (32B-parameter post-trained model on \~10M curated Q\&A pairs with reinforcement learning from verifiable biological rewards). These models are called by skills encoding specialized know-how — for example, how to orchestrate relevant data and tools to produce a target characterization package consistent with customer templates. Skills are provided by Owkin and can be customized per team. ### Outcomes K Pro is designed to improve For target prioritization specifically, K Pro is designed to improve several measurable outcomes: * **Time to first target dossier** (literature + expression + competitive landscape): 2–4 weeks of analyst time before K Pro, vs. hours with K Pro, reviewed and refined collaboratively * **Number of targets evaluated in parallel for a given indication**: a handful before (constrained by analyst capacity), vs. dozens in parallel and consistent campaigns with K Pro * **Reproducibility of scoring across analysts and projects**: same skill applied consistently with versioned cohort definitions * **Evidence coverage (modalities per target)**: often bulk RNA-seq plus literature before, vs. bulk + single-cell + spatial RNA + pathology + clinical landscape in addition to literature with K Pro --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.owkin.com/getting-started/why-k-pro.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.