Lung tumor discrimination by deep neural network model CanDo via DNA methylation in bronchial lavage
Description
The dataset includes raw data from targeted methylation sequencing, as well as the count data and codes used in constructing and validating the CanDo model. Raw data.rar archive contains raw data generated by next-generation sequencing platforms in this study, formatted as FASTQ. These data are organized within subdirectories named after each patient's ID. Extended Data 1.xlsx (EXdata1) includes targeted methylation detection results for 118 patients. Extended Data 2.xlsx (EXdata2) displays the calculated average methylation level of each gene for the 118 patients. Extended Data 3.xlsx (EXdata3) includes targeted methylation detection results for 33 patients in a strictly unbiased validation cohort. Extended Data 4.xlsx (EXdata4) exhibits the calculated average methylation level of each gene for the 33 patients in the validation cohort. Code.docx contains all the code utilized in constructing and validating the CanDo model.