Getting Started with iFeature
Step 1. Paste protein sequences or upload protein sequence file.
|First, paste the protein (or peptide) sequences in 'TEXTAREA' or upload the file, which include the protein sequence. The protein sequences must be in 'FASTA' format. iFeatureWeb was designed to accept at most 100 sequence at once.|
|The following descriptors could be only used to calculate the peptides with equal length:|
|Enhanced Amino Acid Comporition Enhanced Grouped Amino Acid Composition Binary BLOSUM62 Descriptor AAindex Descriptor Z-Scale Descriptor|
Step 2. Select the descriptor(s).
|Selcect the descriptor types, at least one descriptor shoudld be chosen. Then, click the 'Submit' to calculate the selected descriptor(s) or go on to select Clustering Algorithms to cluster samples or descriptors.|
Step 3. Select the clustering algorithms.
|At this step, iFeature allows users to select the clustering algorithm and perform feature clustering. User can select either 'sample clustering' or 'feature clustering' mode. The former is for the purpose of clustering for the input protein or peptide sequences, while the latter is clustering for the extracted feature descriptors. Five commonly used clustering are available.|
Step 4. Select the feature selection algorithms.
|At this step, with the additional 'sample label (see example)' information, iFeature can identify the most characterizing features using the corresponding feature selection algorithm. Four commonly used feature selection algorithms are provided by iFeature.
Finally, clicking 'Submit' will enable users to calculate the feature and perform feature clustering and selection using the selected clustering and selection algorithm.
Step 5. Result.
|After a few seconds, the feature descriptors will be generated and displayed in the result table, where rows represents the submitted protein/peptide sequences, columns represent the derived feature descriptors. The descriptors, clustering and feature selection result can also be downloaded to local computers for further analysis.|
Backend computation is powered by our Python package iFeature.