1. First, insert sequences as text or file, in FASTA format. You can paste your sequences, or upload a FASTA file.
You can then choose to filter out all the sequences that don't have signal peptide (see wiki ). Note that activating the signal peptide filter may slow the process.

2. After submission, the results summary section should appear.
Here, you can view the number of submitted sequences filtered by predicted confidence. "Full confidence" mean that 4 predictor out of 4 predict the sequence as a neuropeptide-precursor. Sequences with illegal amino-acids (i.e. B, X, U, Z) are filtered out.
You can view the full results set ("Detailed Results" tab) - a table with the full classifications for positively predicted proteins and "high quality" predictions - the set of sequences that pass an internal score threshold (see FAQ page).

3. In the "Download / Export" tab you can download sequences according to the classification of sequences to 4 groups, organized by confidence level. You can download the sequences' ids or fasta, or you can download prediction scores in excel format. The is also an option to forward some of the sequences to Neuropred for cleavage products' predictions.
You can choose to get the proteins with high confidence (all four classifiers classify as positive), or less certain proteins (with partial levels of predictor consensus).

4. You can also view the histograms of 12 different features according to known positive (neuropeptides) and negative proteins. Additionally, you may pick individual predicted sequences to see their features' values.

5. The "Sequence Graphics" tab presents a graphic view of the predicted NPPs-sequences. For each protein, a number of features are displayed, with an additional graphical motif of lone, basic (K,R) amino-acids marked by green lines and the dibasic or tribasic amino acid pairs in red. Columns are sortable (As are other tables in NeuroPID.

© 2013 Neuropid | The Hebrew University of Jerusalem | SK