Genome Prediction

Explore TF binding affinity across an arbitrary E. coli genomic region.

1. Genomic Region *

Coordinates are 1-indexed base-pair positions on the E. coli K-12 MG1655 chromosome (NC_000913.3). Region size: 25 – 1,500 bp. Try PdhR example →
?
Valid range: 1 – 4,641,652.
Must be less than the stop coordinate.
?
Valid range: 1 – 4,641,652.
Must be greater than the start coordinate.
Or look up a gene to auto-fill coordinates:
?
Type a gene name to automatically fill the start and stop coordinates with a 1,500 bp window centred on the gene's annotated start site.
Coordinates are derived from the NCBI RefSeq annotation of the E. coli K-12 MG1655 genome (NC_000913.3).
You can adjust the coordinates manually after auto-fill.

2. Output Options

?
How TFs are ranked for this region. The same ranking is used to choose which TFs the plot shows (when no TF models are specified) and to order the score table, so the plot and table always match.

Max binding site score
Strength of the single strongest binding position anywhere in the region.

Full-sequence score
Total binding summed across every position in the region, on both strands.
Applies to both the plotted TFs and the ranked score table.
?
Start typing a TF name and press Enter to add it.
You can add up to 10 TFs; each appears as a removable tag.
Leave empty to automatically show the top-scoring TFs (up to Max TFs displayed).
If TFs are selected, BoltzNet plots only those TFs.
BoltzNet includes 124 TF models. Leave empty to automatically show the top-ranked TFs. Enter TF names to plot only those, in the order provided.
?
Maximum number of TFs to display in the plot.
Accepts values between 1 and 10.
TFs are ranked by predicted binding affinity.
Only used when no TF models are selected.
?
Resolution of the output image.
200 dpi — standard screen quality, smaller file.
300 dpi — high quality (default).
600 dpi — publication quality, larger file.
?
What is the score?
BoltzNet scans every position across both strands of the region and computes a binding energy for each TF at each position. The score for a TF is the sum of binding probabilities across all positions — higher means stronger predicted binding in this region.
Results are displayed as a ranked table of all 124 TFs on the results page, and are also available to download as a CSV file.
?
Limit scores to a specific set of TFs.
Type a TF name and press Enter to add it. Each TF appears as a removable tag.
Leave empty to get scores for all 124 TF models.
Leave empty to score all 124 TFs.

3. Notification Email *

?
When the job is finished, a link to the results page will be sent to this address.
Please enter a valid email address (e.g., name@example.com).
Note: The first genome job will take several minutes while BoltzNet builds the genomewide prediction model. All subsequent jobs will be much faster. The same applies after a BoltzNet version upgrade — the model will be rebuilt once automatically.
Cancel