Tuning the mechanical properties of metals, including strength, through adjusting the type and/or concentration of added solute elements, has been recognized as an effective way to design and produce materials with desired or optimized mechanical properties. Developing predictive models that connect the material properties at atomic level to the macroscopic strength is thus crucial for theory-guided design of new materials with superior mechanical performance. Solute strengthening refers to the additional strength which arises from the totality of the interaction energies between the solutes and an individual dislocation. Prevailing theories for strengthening in body-centered cubic (BCC) alloys consider only solute interactions in the core of the screw dislocation while computations suggest longer-range interactions. In this work, we define the proper solute/screw interaction energy parameter relevant for strengthening of screw dislocations in random bcc alloys from dilute binary alloys to high-entropy alloys. Solutes could be added in small amounts to the pure base metal to form substitutional dilute alloys. It is well-established that the plastic deformation in dilute BCC alloys is controlled by motion of screw dislocations through thermally-activated double-kink nucleation and migration processes. However, the effect of solutes on these processes is not well-established. Here, we develop theoretical models to predict the barrier associated with each of these processes which ultimately enable us to compute the strength of the alloy without any fitting parameters. High-entropy alloys (HEAs) are a new class of random multi-component alloys with impressive mechanical properties. Recent theory suggested that the underlying mechanisms involving the screw dislocation motion in BCC non-dilute and HEAs differs fundamentally from that of dilute alloys and is controlled by a combination of Peierls-like motion, kink migration, and cross-kink failure. Existing kink migration models, in spite of successes in capturing some experiments, are based on several invalid assumptions. Here, a new theory for the kink migration in HEAs is developed based on our recent understanding in dilute alloys, leading to a fully derived analytical model for the kink migration energy barrier. The BCC refractory HEAs composed of the family of Mo-W-V-Nb-Ta are of particular interest due to their high-temperature strength retention. Very recent theoretical and experimental studies have proposed that yield strength in these and other BCC HEAs is controlled by edge dislocation. In this study, the very high energy barriers hindering the edge dislocation is analyzed using atomistic simulations, leading to high strengths and high strength retention at elevated temperatures. Many BCC refractory HEAs show a distinct plateau in strength versus temperature at intermediate temperatures. In the last part of this thesis, we examine one possible mechanism for the intermediate-T strength plateau: the dynamic strain aging (DSA) process of solute diffusion immediately across the core of an edge dislocation. An analytic model is developed which captures the major dependencies in terms of underlying material properties, and can thus be applied to other alloys. The predictive theoretical models developed in this thesis pave the way for theory-guided design of novel high-performance materials with excellent or even unprecedented mechanical properties.