2 results about "Differential effects" patented technology

The present invention discloses a method and system for discovering single-cell drug causal genes based on adversarial transfer learning, which relates to the technical field of drug sensitivity prediction, including: obtaining source domain data sets and target domain data sets, constructing an adversarial transfer model, training and optimizing the adversarial transfer model according to the source domain data sets and the target domain data sets, and obtaining single-cell drug sensitivity prediction results; modeling single-cell gene expression data, combining the k-nearest neighbor matching algorithm with the single-cell drug sensitivity prediction results, obtaining counterfactual values, and then calculating confounding effects and differential effects; calculating causal effects based on confounding effects and differential effects, and screening out causal genes of single-cell drugs. On the one hand, the present invention solves the key problem of lack of drug sensitivity annotation data in single-cell analysis, and on the other hand, successfully discovers causal genes, providing important clues for in-depth understanding of the drug action mechanism.

Each atom in a population of atoms can be characterized by a probability density distribution (PDD). Using a shaken-lattice technique, each PDD is split into a pair of sub-PDDs. The sub-PDDs of a pair are propagated along different paths to a common endpoint of the paths, resulting in a matter-wave interference pattern that encodes a net phase between the paths, e.g., due to differential effects associated with a gravity gradient. The matter-wave interference pattern can be measured to yield a respective measurement for each atom. The measurements can be aggregated to yield a result distribution that can serve as a classical domain estimate of the quantum-domain matter-wave interference pattern, and thus of the gravity gradient. Other embodiments can determine gradients for other types of fields.