pubmed.ncbi.nlm.nih.gov

CHEXVIS: a tool for molecular channel extraction and visualization - PubMed

  • ️Thu Jan 01 2015

CHEXVIS: a tool for molecular channel extraction and visualization

Talha Bin Masood et al. BMC Bioinformatics. 2015.

Abstract

Background: Understanding channel structures that lead to active sites or traverse the molecule is important in the study of molecular functions such as ion, ligand, and small molecule transport. Efficient methods for extracting, storing, and analyzing protein channels are required to support such studies. Further, there is a need for an integrated framework that supports computation of the channels, interactive exploration of their structure, and detailed visual analysis of their properties.

Results: We describe a method for molecular channel extraction based on the alpha complex representation. The method computes geometrically feasible channels, stores both the volume occupied by the channel and its centerline in a unified representation, and reports significant channels. The representation also supports efficient computation of channel profiles that help understand channel properties. We describe methods for effective visualization of the channels and their profiles. These methods and the visual analysis framework are implemented in a software tool, CHEXVIS. We apply the method on a number of known channel containing proteins to extract pore features. Results from these experiments on several proteins show that CHEXVIS performance is comparable to, and in some cases, better than existing channel extraction techniques. Using several case studies, we demonstrate how CHEXVIS can be used to study channels, extract their properties and gain insights into molecular function.

Conclusion: CHEXVIS supports the visual exploration of multiple channels together with their geometric and physico-chemical properties thereby enabling the understanding of the basic biology of transport through protein channels. The CHEXVIS web-server is freely available at http://vgl.serc.iisc.ernet.in/chexvis/ . The web-server is supported on all modern browsers with latest Java plug-in.

PubMed Disclaimer

Figures

Figure 1
Figure 1

2D illustration of an alpha complex based representation of a molecule and the empty space within.(a) Union of disks (balls in 3D represent atoms) where the contribution from each disk is equal to its intersection with the corresponding Voronoi cell. (b) The weighted Delaunay triangulation of the disks and the convex hull (green). (c) Alpha complex at α=0, shown in red, is a subcomplex of the weighted Delaunay triangulation. (d) A cavity is a connected component of the complement of the alpha complex. A cavity with atleast one opening is a pocket (blue), while buried cavities are referred to as voids (green). (e) The empty space represented by the cavity triangles. (f) A channel is a simply connected subset of simplices of a pocket each of whose triangles has at most two neighbors and at least one boundary edge is a mouth edge. Here a pore (pink), a channel with two openings, is shown represented as a subset of the complement of the alpha complex. (g) A channel from the boundary to an interior point. (h) Underlying empty space of the channel. (i) Simplices of the complement of the alpha complex that represent the channel. (j) A path representation of the channel in which nodes are located at the centers of the orthogonal circle corresponding to each triangle and arcs connect nodes that correspond to neighbouring triangles.

Figure 2
Figure 2

Channel extraction in a transmembrane protein.(a) Mechanosensitive Channel of Large Conductance (MscL, PDB id 2OAR). (b) The path network. (c) Widest path tree. (d) Pruned widest path tree. (e) Pores between top 10 boundary nodes in the network. (f) Transmembrane pores between top 5 interior and 5 exterior nodes. (g) Top ranked transmembrane pore shown using the skin surface (yellow). Other pores are shown for context.

Figure 3
Figure 3

Channel profile visualization of the transmembrane pore in 2BG9.(a) Conservation profile shown as a color map over the radius profile, which in turn is shown as a symmetric graph plot. (b) Hydrophobicity and conservation profiles shown in a split visualization over the radius profile.

Figure 4
Figure 4

Illustration of 2D representation of a channel in a synthetic 2D example.Top: The molecule is shown as a set of disks (grey). The channel is shown as a path (red) and as a set of triangles (blue). In 2D, each node in the path corresponds to a triangle, while a path edge corresponds to an edge in the triangulation. Each disk is given an alphabetic label, while the nodes of the path are given numeric labels. Bottom: The proposed 2D visual metaphor of the channel is shown. Each vertical box denotes a node in the path (and thus denotes a triangle within the channel). The three small boxes within the node box denote the disks incident on the triangle corresponding to the particular node box. Also, consecutive nodes boxes are connected by two edges. These edges denote the edge shared by the consecutive triangles in the channel. For example, node boxes 3 and 4 denote triangles cbf and cgf, respectively. Their shared edge is cf, so the atom boxes corresponding to discs c and f are connected by edges. This representation naturally extends to the 3D case, with the only modification that each node box will denote a tetrahedron and will thus contain four atom boxes. Two consecutive node boxes will be connected by three edges denoting the common triangle shared by the adjacent tetrahedra.

Figure 5
Figure 5

2D representation of a channel.(a) 2D box representation of the atoms lining a portion of the channel in 2BG9. (b, c) Boxes are merged and colored by atom type and polarity of residues. The numbers below the boxes are tetrahedron indices, which makes it clear that this visualization corresponds to a subset (tetrahedra 46-56) of the complete channel.

Figure 6
Figure 6

Comparison of channel extraction tools. Results obtained from different channel extraction tools, viz.

CHEXVIS

,

MOLE

,

CAVER

,

MOLAXIS

and

POREWALKER

, are summarized in a tabular form. On the top, results for some enzymes with active sites are shown while results for transmembrane are shown on the rows at the bottom. For complete comparison results, refer to Additional files 3 and 4.

Figure 7
Figure 7

Comparison of pores extracted by

MOLE

and

CHEXVIS

in 3EAM.(a) Cartoon representation of the transmembrane protein 3EAM. The 3D view is such that the transmembrane pore is perpendicular to the plane of the page. (b) The pores extracted by

MOLE

in this structure.

MOLE

identifies a few side channels going from central pore to outside, but it fails to identify the transmembrane pore. (c) On the other hand,

CHEXVIS

identifies the main transmembrane pore as well as some side channels. (d) Also, the top transmembrane pore suggested by

CHEXVIS

is verified to be the correct transmembrane pore in this structure [39].

Figure 8
Figure 8

Comparison of pores extracted by

MOLE

,

MOLAXIS

,

POREWALKER

and

CHEXVIS

in 2J1N.(a) Cartoon representation of the transmembrane protein 2J1N. This structure consists of three beta-barrel subunits going across the cell membrane. The figure shows top view of the protein such that cell membrane is parallel to the plane of the page. (b)

MOLE

identifies two pores in this structure, both of them are not correct transmembrane pore passing through beta-barrels.

MOLE

wrongly identifies the narrow space between the three units as a transmembrane pore. (c)

MOLAXIS

also identifies a pore going through space between three subunits. This maybe due to the specified input parameters. (d)

POREWALKER

also fails to identify the correct transmembrane pores in this structure. It identifies the empty space between three subunits as the only transmembrane pore. By design,

POREWALKER

would not have been able to identify all the three transmembrane pores as it extracts only one transmembrane pore. (e)

CHEXVIS

is able to correctly identify transmembrane pores through all the three subunits, using default parameters for finding pores. (f) Unlike other tools which identify pore through the space between subunits, the top transmembrane pore identified by

CHEXVIS

is one of the pore passing through a beta-barrel subunit.

Figure 9
Figure 9

The pentameric ligand-gated ion channels extracted in closed and open conformations. The channels are wide and hydrophilic in extra-cellular region which is less conserved. The constricted transmembrane spanning helical channel section is hydrophobic and highly conserved. The bottleneck radius is 2.21Å in closed configuration while it is 5.12Å in open state.

Figure 10
Figure 10

Transient Receptor Potential (TRP) channels extracted in closed and open state. The constricted region of the channel (highlighted by rectangles) is both highly conserved and hydrophobic. The bottleneck radius is 2.6Å in closed configuration while it is 4.66Å in open state.

Figure 11
Figure 11

Outer Membrane (OM) carboxylate channels in structures 3SYS and 3SY7. These channels are narrower compared to channel porin (2J1N) indicating that they are more selective.

Figure 12
Figure 12

Properties of membrane carboxylate channel in 3SYS. The 2D profile is coloured by conservation and hydrophobicity of residues. It can be observed that the channel constriction is more conserved than the rest of the channel. According to box representation of first row, the amino acids lining the channel shows higher proportion of Arginine. Most of the residue side chains point towards the channel as can be concluded by many red boxes in the second row. The third row shows that channel is surrounded by loops (coloured yellow), specially at the constriction. Although some residues at channel end points belong to outer beta-barrel structure. Lastly, the fourth row shows the chemical properties of the residues lining the channel. The basic residues which play an important role in the function of this channel are correctly identified by ChExVis. They are represented as dark blue boxes.

Figure 13
Figure 13

The KcsA potassium channels extracted in structures 1K4C, 1K4D and 1S5H. On left, channel extracted in 1K4C (high K+ concentration) is shown. The channel has four highly conserved K+ sites which are surrounded by carbonyl Oxygens as shown in the profile shown. The channel closes in low K+ concentration (1K4D) as captured by next profile. The channel in this structure is more constricted and one of the site is surrounded by Carbon atoms instead of carbonyl Oxygens. Lastly, on right a mutant channel (1S5H) is shown which has reduced K+ conduction capability. This is attributed to replacement of Oxygens with Carbon atoms at crucial site no. 4, which is correctly captured by

CHEXVIS

in the profile shown.

Similar articles

Cited by

References

    1. Zhou HX, McCammon JA. The gates of ion channels and enzymes. Trends Biochem Sci. 2010;35(3):179–85. doi: 10.1016/j.tibs.2009.10.007. - DOI - PMC - PubMed
    1. Hubner CA, Jentsch TJ. Ion channel diseases. Hum Mol Genet. 2002;11(20):2435–45. doi: 10.1093/hmg/11.20.2435. - DOI - PubMed
    1. Sehnal D, Vařeková RS, Berka K, Pravda L, Navratilová V, Banáš P, et al. MOLE 2.0: advanced approach for analysis of biomacromolecular channels. J Cheminform. 2013;5(1):39. doi: 10.1186/1758-2946-5-39. - DOI - PMC - PubMed
    1. Levitt DG, Banaszak LJ. POCKET: a computer graphics method for identifying and displaying protein cavities and their surrounding amino acids. J Mol Graph. 1992;10(4):229–34. doi: 10.1016/0263-7855(92)80074-N. - DOI - PubMed
    1. Hendlich M, Rippmann F, Barnickel G. LIGSITE: automatic and efficient detection of potential small molecule-binding sites in proteins. J Mol Graph Model. 1997;15(6):359–63. doi: 10.1016/S1093-3263(98)00002-3. - DOI - PubMed

Publication types

MeSH terms

Substances