BLISS 2.0

About BLISS 2.0
>> An Introduction to Researchers
>> Tutorial
>> Supplement Information

CRM Analysis
>> Cis-Regulatory Modules(CRMs) Search
>> Single Cis-Regulatory Module(CRM) Search

Publications
>>Publications

Given two orthologous DNA sequences, BLISS 2.0 is able to output all potentially conserved regulatory modules between those two sequences. BLISS 2.0 analysis proceeds in Three Major Steps:

Step 1:

To begin, users are required to input two raw DNA sequences. Due to the limits of CPU and memory of our server, BLISS 2.0 currently can not analyze very long DNA sequences. Please limit the sum of the lengths of the two sequences below 10K bps.

BLISS 2.0 generates binding site profiles for each sequence based on matrices collected by Transfac 9.1. BLISS 2.0 searches binding sites not only in the sequences but also in their complementary sequences.  If the first sequence is from the forward sequence of one gene and the second sequence is from the complementary sequence of the orthologous gene, in which case the order of conserved binding sites are reversed, the user can check the checkbox under the text area of the second sequence to prevent the potential miss because the orientation of the sequences.

And on the same page, users are allowed to choose a binding site score cutoff. Either 0.75 or 0.8 can be selected as the binding score cutoff to determine if a specific binding site is found in a certain location in the DNA sequence. This cutoff is based on a balance between specificity and sensitivity. Higher score cutoff increases the specificity, but decreases the sensitivity at the same time. 

Step 2:

By integrating Gaussian smoothing and statistical analysis on the binding site profiles comparison, BLISS 2.0 indicates the degree of the conservation at the binding site level between two sequences as BLISS2_scores, which are displayed and visualized as a two-dimensional color plot.

The vertical color bar on the right side of the color plot shows the value of BLISS2_score that the color represents. Continuous high BLISS2_scores along diagonal direction indicates a potential match of conserved regulatory modules between two sequences.

There are two BLISS2_score distributions. The left one is the distribution when a random sequence matches with a random sequence or a sequence harboring a regulatory module and the right is the distribution when two conserved regulatory modules (clusters) match. Based on this statistical analysis, users can choose a BLISS2_score cutoff to output all matched regions with the BLISS2_score greater than this cutoff to be conserved regulatory modules (clusters).

Binding sites in the matched regions are going to be displayed. There is a balance between sensitivity and specificity when choosing a cutoff to output binding sites. BLISS 2.0 allows the user to choose either run the program for “High sensitivity” (p value < 0.01) or “High specificity” (p value < 0.001). Users also have the option to determine how to rank the shared binding sites in reported regulatory modules. They can be ranked by locations, by numeric contributions to BLISS2_score or by product of p-values of the matching binding sites on both sequences. 

Step 3:

Finally, all the matching regulatory modules (clusters) with BLISS2_score greater than the BLISS2_score cutoff that users choose are displayed on the third page.

Contributing binding sites are listed in a separated table for each matched region. Based on the choice of users, binding sites can be ranked by locations, by numeric contributions to BLISS2_score or by product of p-values of the matching binding sites on both sequences. 

To help users identify/select binding sites that are more likely to be functionally important or pertinent to their research interest, BLISS 2.0 provides the option to let users highlight the binding sites they are interested in as green color.

Single Cis-Regulatory Module(CRM) Search

It has been experienced by many researchers that the laboratory experiments in a model organism has narrowed down the location of the regulatory module for a particular gene to a relatively short region, whereas for the ortholog in a less-studied organism, information about the localization of the module is absent. To be able to deal with this special instance, BLISS 2.0 provide a complementary tool – Single Cis-Regulatory Module Search, to locate the position of the conserved regulatory module in the ortholog of the less-studied organism. We suggest users to use this tool especially in the case that the length of one of the orthologous DNA sequences is less than 200 bps. The procedure of this analysis is very similar with that of Cis-Regulatory Modules(CRMs) Search as explained above.