MUMBAI, India, Aug. 17 -- Intellectual Property India has published a patent application (202641095788 A) filed by Amrita Vishwa Vidyapeetham on August 07, 2026, for A Method For Generating A Probability-Based Classification Output From Gut Microbiome Sequencing Data.
Inventors include Babu, Shekar; Govindarajan, Priya G; Ravi, Harshitha; and Ajjenahalli Kalegowda, Anusha.
The application for the patent was published on August 14, 2026, under issue no. 33/2026.
Abstract: The present invention relates to a method for generating a probability-based classification output from gut microbiome sequencing data. A pre-generated microbiome feature table is filtered by taxa prevalence and transformed into a compositional log-ratio representation correcting a constant-sum constraint. A subset of features is selected using a mutual-information score, and base learners modelling non-linear and linear relationships are trained across stratified cross-validation folds to generate out-of-fold probability outputs. A meta-learner combines these outputs, and training is repeated across randomized partitions and aggregated into an ensemble- aggregated probability score, from which a numeric value and associated data label are generated. Figure 1 is the representative figure.
Disclaimer: Curated by HT Syndication.
