Systems and methods for analyzing omics data

A method of correcting batch effects in proteomic data, comprising:

• (a) providing a neural network comprising:
• i. an input layer configured to receive at least one polyamino acid descriptor;
• ii. a latent layer configured to output at least a latent descriptor, wherein the latent layer is operably connected to the input layer;
• and
• iii. an output layer operably connected to the latent layer;
• wherein the latent layer and the output layer comprises one or more parameters;
• (b) providing training data comprising a plurality of the at least one polyamino acid descriptors, wherein the plurality of polyamino acid descriptors comprises at least one value for a measured intensity of a given polyamino acid;
• (c) training the neural network, by (i) inputting at least the plurality of polyamino acid descriptors at the input layer of the neural network, (ii) outputting a plurality of latent descriptors at the latent layer and a plurality of outputs at the output layer, and (iii) optimizing a loss function that is configured to guide the neural network towards learning a latent space comprising a plurality of embeddings for the plurality of polyamino acid descriptors by updating the one or more parameters, wherein the plurality of embeddings is invariant with respect to a predetermined non-biological factor.

A method of correcting batch effects in omic data, comprising:

• (a) providing a neural network comprising:
• i. an input layer configured to receive at least one omic descriptor;
• ii. a latent layer configured to output at least a latent descriptor, wherein the latent layer is operably connected to the input layer;
• and
• iii. an output layer operably connected to the latent layer;
• wherein the latent layer and the output layer comprises one or more parameters;
• (b) providing training data comprising a plurality of the at least one omic descriptors wherein the plurality of omic descriptors comprises at least one value for a measured intensity of a given omic signal;
• and
• (c) training the neural network, by (i) inputting at least the plurality of omic descriptors at the input layer of the neural network, (ii) outputting a plurality of latent descriptors at the latent layer and a plurality of outputs at the output layer, and (iii) optimizing a loss function that is configured to guide the neural network towards learning a latent space comprising a plurality of embeddings for the plurality of omic descriptors by updating the one or more parameters, wherein the plurality of embeddings is invariant with respect to a predetermined non-biological factor.