The Colosseum Open RAN Digital Twin is a world-leading wireless network emulator with hardware-in-the-loop capabilities, designed to bridge the gap between the Open RAN vision and its deployment and commercialization. This paper presents Colosseum as a tutorial on how it can provide the research infrastructure and tools needed to address the challenges of Open RAN development. Colosseum implements an Open RAN digital twin through a high-fidelity Radio Frequency (RF) channel emulator and end-to-end softwarized O-RAN and 5G-compliant protocol stacks, allowing users to reproduce and experiment upon topologies representative of real-world cellular deployments. The paper details the twinning infrastructure of Colosseum, as well as the automation pipelines for RF and protocol stack twinning. It showcases a broad range of Open RAN use cases implemented on Colosseum, including real-time connection between the digital twin and real-world networks, and the development, prototyping, and testing of AI/ML solutions for Open RAN. The Open RAN paradigm is transforming the deployment, management, and optimization of cellular systems by introducing openness, softwarization, programmability, interoperability, and intelligence. However, advancing the Open RAN vision requires addressing several challenges, including the need for datasets to train AI/ML models, end-to-end AI/ML testing, continuous software validation, and automated integration and testing of disaggregated components. Colosseum addresses these challenges by providing a digital twin platform that allows for the replication of real-world scenarios, the generation of datasets for AI/ML training, and the development of end-to-end, fully integrated, and reliable solutions for Open RAN. Colosseum is a wireless network emulator with hardware-in-the-loop capabilities, consisting of 128 pairs of generic compute servers and SDRs, a channel emulation system, and a state-of-the-art AI/ML infrastructure. It enables the replication of real-world RF scenarios and the development of open-source cellular protocol stacks, such as OAI for 5G RANs and srsRAN for 4G networks. The OpenRAN Gym framework is used to instantiate softwarized RANs based on these protocol stacks and control them through xApps deployed on an O-RAN Near-Real-time (RT) RIC. The framework allows users to develop and test AI/ML solutions for Open RAN, including DRL agents for optimizing RAN performance. Colosseum also supports real-time RF twinning through a combination of real-time channel modeling and sensing techniques. The channel emulation scenario generator and sounder toolchain (CaST) enables the generation of digital twin RF scenarios by pre-generating offline scenarios and replaying them in real-time. The CaST framework allows users to characterize real-world RF environments and turn them into digital twin representations for use in channel emulators such as Colosseum. The framework includes a streamlinedThe Colosseum Open RAN Digital Twin is a world-leading wireless network emulator with hardware-in-the-loop capabilities, designed to bridge the gap between the Open RAN vision and its deployment and commercialization. This paper presents Colosseum as a tutorial on how it can provide the research infrastructure and tools needed to address the challenges of Open RAN development. Colosseum implements an Open RAN digital twin through a high-fidelity Radio Frequency (RF) channel emulator and end-to-end softwarized O-RAN and 5G-compliant protocol stacks, allowing users to reproduce and experiment upon topologies representative of real-world cellular deployments. The paper details the twinning infrastructure of Colosseum, as well as the automation pipelines for RF and protocol stack twinning. It showcases a broad range of Open RAN use cases implemented on Colosseum, including real-time connection between the digital twin and real-world networks, and the development, prototyping, and testing of AI/ML solutions for Open RAN. The Open RAN paradigm is transforming the deployment, management, and optimization of cellular systems by introducing openness, softwarization, programmability, interoperability, and intelligence. However, advancing the Open RAN vision requires addressing several challenges, including the need for datasets to train AI/ML models, end-to-end AI/ML testing, continuous software validation, and automated integration and testing of disaggregated components. Colosseum addresses these challenges by providing a digital twin platform that allows for the replication of real-world scenarios, the generation of datasets for AI/ML training, and the development of end-to-end, fully integrated, and reliable solutions for Open RAN. Colosseum is a wireless network emulator with hardware-in-the-loop capabilities, consisting of 128 pairs of generic compute servers and SDRs, a channel emulation system, and a state-of-the-art AI/ML infrastructure. It enables the replication of real-world RF scenarios and the development of open-source cellular protocol stacks, such as OAI for 5G RANs and srsRAN for 4G networks. The OpenRAN Gym framework is used to instantiate softwarized RANs based on these protocol stacks and control them through xApps deployed on an O-RAN Near-Real-time (RT) RIC. The framework allows users to develop and test AI/ML solutions for Open RAN, including DRL agents for optimizing RAN performance. Colosseum also supports real-time RF twinning through a combination of real-time channel modeling and sensing techniques. The channel emulation scenario generator and sounder toolchain (CaST) enables the generation of digital twin RF scenarios by pre-generating offline scenarios and replaying them in real-time. The CaST framework allows users to characterize real-world RF environments and turn them into digital twin representations for use in channel emulators such as Colosseum. The framework includes a streamlined