Dealing with duplicate records

Sample duplicates:

This utility function is used to process the estimation on the technical replicates with different mathematical models; Samples of the replicates should be formatted with underscore and integer, e.g. SampleA_1. You can see an example technical replicates here.

• Duplicates values can be merged by estimating with simple arithmetic mean, minimum, maximum, median, sum, or quantile.
• When many replicates (at least 3) are provided, you can also smooth values based on kernel density estimation (Gaussian kernel, bandwidth 0.5) before merging
• If a feature shows over 1/3 missing and its CV is > 1, the feature in the corresponding sample will be assigned as 0;

Feature duplicates:

This utility function is used to merge duplicate features in putative annotated results (such as by mzcloud or chemspider) in untargeted metabolomics. Data containing feature replicates must be uploaded in the format of Samples in columns. You can see an example data containing feature duplicates here.

• Duplicate values can be replaced by their simple arithmetic mean, median, sum, minimum, maximum, or quantile. In general, only the first two options are recommended
• Missing values will be ignored

Please upload your data table