DispHScan is a web tool designed to predict protein disorder as a function of pH for multiple sequences. This new functionality offers the possibility to conduct pH-dependent disorder analysis at the proteome-wide level that might reveal new insights into the physiological and pathological role of the solution pH in the conformational plasticity of IDPs and assist scientists and industries in identifying optimal pH conditions for proteins of interest. DispHScan is freely available for academic users at: http://disphscan.ppmclab.com/
1. Introduction
Intrinsically disordered proteins (IDPs) do not have a static tertiary structure but fluctuate between unfolded and partially folded states. This conformational plasticity is fundamental for their biological activity as regulatory elements in the cell
[1]. The unfolded nature of IDPs is encoded in their primary structure
[2] and, therefore, can be predicted from the polypeptide sequence. Multiple computational tools, exploiting diverse physicochemical protein principles, have succeeded in identifying disordered proteins in standard conditions
[3], yet without considering environmental variables such as pH in their algorithms. In a recent study, we modeled protein disorder as a function of pH and developed a disorder predictor able to anticipate disorder-to-order transitions in IDPs named DispHred
[4]. Still, DispHred was limited to the prediction of a unique polypeptide sequence at a time, precluding the study in proteomes. With DispHScan we addressed this need providing a new tool that expanded the capabilities of our original method allowing the study of pH-dependent protein disorder for multiple sequences.
2. DispHScan Pipeline
DispHScan outlines pH-dependent disorder in a defined pH interval and identifies conformational transitions for multiple sequences. The server computes the charge and hydrophobicity as a function of pH and applies a linear boundary condition to discriminate the folding state of each individual sequence (Figure 1).
Figure 1. DispHScan pipeline. Users must introduce their sequences in FASTA format and select the pH interval with the desired step and window sizes (default values are 0.5 and 51, respectively). The option of predicting disorder at a single pH is also available. The server computes mean hydrophobicity and NCPR to provide a disorder prediction for each sequence and pH. Possible transitions are checked in the interval of study. The results are represented in both tabular and graphical formats, as well as in JSON; all of them available for download.
3. Performance
DispHScan was tested on the human proteome (UP000005640) from pH 0 to pH 14 using a step size of 0.5 units and a window size of 51 residues. The program was able to scan a total of 20,600 sequences, each at 29 different pHs (597,000 data points) in less than 17 h, rendering a proportion of 6.4% transitioning proteins.
Besides, we conducted a pH-dependent disorder analysis in the proteomes of three additional model organisms, including Escherichia coli (UP000000558), Saccharomyces cerevisiae (UP000002311), and Caenorhabditis elegans (UP000001940).
4. Future Perspectives
Considering the novelties of this tool, we expect that DispHScan would find application in investigating the role of pH-conditioned disorder in different aspects of cell biology at a proteome-wide level. DispHScan might also be of interest for biotechnological applications by helping researchers detect stable conformations at pH values far from neutrality, or by assisting in the redesign of protein variants displaying a conformational response to changes in the solution pH. It may also help in assessing pH-dependent disorder for optimizing solvent conditions and to speed up buffer screenings.