NPLB

No Promoter Left Behind

NPLB finds promoter architectures (PAs) and their corresponding promoter elements (PEs) from a given set of promoter sequences. It is available both as a web based application and as a downloadable software.

Help
NPLB software comes with two functionalities.

promoterLearn

To fill in the details of executing promoterLearn, select the first radio button
Input a valid FASTA file. This is the only compulsory input for running promoterLearn. (sample)
Minimum number of PAs possible for given dataset. Default value: 1
Maximum number of PAs possible for given dataset. Default value: 20
Sort PAs in increasing order of median values calculated from values in given column of additional file. It is compulsory to provide a valid additional file and a column number to plot, in order to sort the PAs
Plot data in given column of the additional file in the form of pie charts or boxplots with respect to the PAs of the best model
A tab separated file consisting of information related to each sequence of the input FASTA file. (sample)
Enter your name
Enter a valid email address. This field is not compulsory. On providing a valid email address, a copy of the results would be mailed
Choose between varying and specific. On selecting varying, models would be learned by varying lambda value and the one with the best cross validation likelihood would be chosen. Default: Varying
Number of models to be learned while training. Only the best model is considered. Default value: 5
Click 'Yes' to save likelihood plots of every model learned. Default: likelihood plots are not saved
Number of folds for K-fold cross validation. Default value: 5
Reset all fields in the form
Submit for execution
Upon submission of the form, a link would be provided. This link displays the status of execution. The status can be QUEUED, RUNNING or COMPLETE. When complete, the results can be downloaded. The results would be saved for 48 hours after execution is over.

promoterClassify

To fill in the details of executing promoterClassify, select the second radio button
A tab separated file consisting of information related to each sequence of the input FASTA file. (sample)
Input a valid FASTA file. This is a compulsory input for running promoterClassify. (sample)
Input a valid model file. This is a compulsory input for running promoterClassify. (sample)
Plot data in given column of the additional file in the form of pie charts or boxplots with respect to the PAs of the best model
Enter your name
Sort PAs in increasing order of median values calculated from values in given column of additional file. It is compulsory to provide a valid additional file and a column number to plot, in order to sort the PAs
Enter a valid email address. This field is not compulsory. On providing a valid email address, a copy of the results would be mailed
Reset all fields in the form
Submit for execution
Note: A valid model file is a binary file (called bestModel.p) produced on running promoterLearn. It is valid only if the length of the sequences in the given FASTA file is same as the length of sequences in the FASTA file on which promoterLearn was executed in order to produce the given model file.

Sequences in FASTA files cannot be more than 200 nucleotides long. Number of sequences cannot be less than 20 and more than 2000. In order to execute large datasets, NPLB can be downloaded and installed on any Linux and Mac systems.

Input files (both FASTA and additional file) cannot be greater than 2MB.

Upon submission of the form, a link would be provided. This link displays the status of execution. The status can be QUEUED or COMPLETE. When complete, the results can be downloaded. The results would be saved for 48 hours after execution is over.