membrane topology

Topology of a transmembrane protein refers to locations of N- and C-termini of membrane-spanning polypeptide chain with respect to the inner or outer sides of the biological membrane occupied by the protein.{{cite journal |last1=von Heijne |first1=Gunnar |s2cid=22218266 |title=Membrane-protein topology |journal=Nature Reviews Molecular Cell Biology |date=December 2006 |volume=7 |issue=12 |pages=909–918 |doi=10.1038/nrm2063 |pmid=17139331 }}

File:Group 1 and 2 transmembrane protein.png

Several databases provide experimentally determined topologies of membrane proteins. They include Uniprot, TOPDB,{{cite journal |last1=Tusnády |first1=Gábor E. |last2=Kalmár |first2=Lajos |last3=Simon |first3=István |title=TOPDB: topology data bank of transmembrane proteins |journal=Nucleic Acids Research |date=1 January 2008 |volume=36 |issue=suppl_1 |pages=D234–D239 |doi=10.1093/nar/gkm751 |pmid=17921502 |pmc=2238857 |doi-access=free }}{{cite journal |last1=Dobson |first1=László |last2=Langó |first2=Tamás |last3=Reményi |first3=István |last4=Tusnády |first4=Gábor E. |title=Expediting topology data gathering for the TOPDB database |journal=Nucleic Acids Research |date=28 January 2015 |volume=43 |issue=D1 |pages=D283–D289 |doi=10.1093/nar/gku1119 |pmid=25392424 |pmc=4383934 |doi-access=free }}{{Cite web|url=http://topdb.enzim.hu/|title=.:TOPDB:.|website=topdb.enzim.hu}} OPM, and ExTopoDB.{{cite journal |last1=Tsaousis |first1=Georgios N. |last2=Tsirigos |first2=Konstantinos D. |last3=Andrianou |first3=Xanthi D. |last4=Liakopoulos |first4=Theodore D. |last5=Bagos |first5=Pantelis G. |last6=Hamodrakas |first6=Stavros J. |title=ExTopoDB: a database of experimentally derived topological models of transmembrane proteins |journal=Bioinformatics |date=1 October 2010 |volume=26 |issue=19 |pages=2490–2492 |doi=10.1093/bioinformatics/btq362 |pmid=20601677 |doi-access=free }}{{Cite web|url=http://bioinformatics.biol.uoa.gr/ExTopoDB/|title=ExTopoDB - Index|website=bioinformatics.biol.uoa.gr}} There is also a database of domains located conservatively on a certain side of membranes, TOPDOM.{{Cite web|url=http://topdom.enzim.hu/|title=.:TOPDOM:.|website=topdom.enzim.hu}}

Several computational methods were developed, with a limited success, for predicting transmembrane alpha-helices and their topology. Pioneer methods utilized the fact that membrane-spanning regions contain more hydrophobic residues than other parts of the protein, however applying different hydrophobic scales altered the prediction results. Later, several statistical methods were developed to improve the topography prediction and a special alignment method was introduced.{{Cite web|url=https://www.sbc.su.se/~miklos/DAS/maindas.html|title=DAS}} According to the positive-inside rule,{{cite journal |last1=von Heijne |first1=Gunnar |title=The distribution of positively charged residues in bacterial inner membrane proteins correlates with the trans-membrane topology |journal=The EMBO Journal |date=November 1986 |volume=5 |issue=11 |pages=3021–3027 |doi=10.1002/j.1460-2075.1986.tb04601.x |pmid=16453726 |pmc=1167256 }} cytosolic loops near the lipid bilayer contain more positively-charged amino acids. Applying this rule resulted in the first topology prediction methods. There is also a negative-outside rule in transmembrane alpha-helices from single-pass proteins, although negatively charged residues are rarer than positively charged residues in transmembrane segments of proteins.{{cite journal |last1=Baker |first1=James Alexander |last2=Wong |first2=Wing-Cheong |last3=Eisenhaber |first3=Birgit |last4=Warwicker |first4=Jim |last5=Eisenhaber |first5=Frank |title=Charged residues next to transmembrane regions revisited: "Positive-inside rule" is complemented by the "negative inside depletion/outside enrichment rule" |journal=BMC Biology |date=2017 |volume=15 |issue=1 |page=66 |doi=10.1186/s12915-017-0404-4 |pmid=28738801 |pmc=5525207 |doi-access=free }} As more structures were determined, machine learning algorithms appeared. Supervised learning methods are trained on a set of experimentally determined structures, however, these methods highly depend on the training set.{{cite journal |last1=Krogh |first1=Anders |last2=Larsson |first2=Björn |last3=von Heijne |first3=Gunnar |last4=Sonnhammer |first4=Erik L.L |s2cid=15769874 |title=Predicting transmembrane protein topology with a hidden markov model: application to complete genomes11Edited by F. Cohen |journal=Journal of Molecular Biology |date=January 2001 |volume=305 |issue=3 |pages=567–580 |doi=10.1006/jmbi.2000.4315 |pmid=11152613 |url=https://pdfs.semanticscholar.org/05e7/f7d5d5a08e0bcaca7497951c9560c546353d.pdf |archive-url=https://web.archive.org/web/20200804085349/https://pdfs.semanticscholar.org/05e7/f7d5d5a08e0bcaca7497951c9560c546353d.pdf |url-status=dead |archive-date=2020-08-04 }}{{Cite web|url=http://www.cbs.dtu.dk/services/TMHMM/|title=TMHMM-2.0 - redirect|website=www.cbs.dtu.dk}}{{Cite web|url=https://phobius.sbc.su.se/|title=Phobius|website=phobius.sbc.su.se}}{{Cite web|url=http://octopus.cbr.su.se/|title=OCTOPUS server}} Unsupervised learning methods are based on the principle that topology depends on the maximum divergence of the amino acid distributions in different structural parts.{{cite journal |last1=Tusnády |first1=Gábor E. |last2=Simon |first2=István |s2cid=15027232 |title=Principles governing amino acid composition of integral membrane proteins: application to topology prediction 1 1Edited by J. Thornton |journal=Journal of Molecular Biology |date=October 1998 |volume=283 |issue=2 |pages=489–506 |doi=10.1006/jmbi.1998.2107 |pmid=9769220 |url=https://pdfs.semanticscholar.org/1556/9480d2117e2b20e6db276de140d19db57831.pdf |archive-url=https://web.archive.org/web/20200208145921/https://pdfs.semanticscholar.org/1556/9480d2117e2b20e6db276de140d19db57831.pdf |url-status=dead |archive-date=2020-02-08 }}{{Cite web|url=http://www.enzim.hu/hmmtop/|title=HMMTOP|website=www.enzim.hu}} It was also shown that locking a segment location based on prior knowledge about the structure improves the prediction accuracy.{{cite journal |last1=Tusnady |first1=G. E. |last2=Simon |first2=I. |title=The HMMTOP transmembrane topology prediction server |journal=Bioinformatics |date=1 September 2001 |volume=17 |issue=9 |pages=849–850 |doi=10.1093/bioinformatics/17.9.849 |pmid=11590105 |doi-access= }} This feature has been added to some of the existing prediction methods. The most recent methods use consensus prediction (i.e. they use several algorithms to determine the final topology) {{Cite web|url=https://topcons.net/|title=TOPCONS: Consensus prediction of membrane protein topology and signal peptides|website=topcons.net}} and automatically incorporate previously determined experimental informations.{{Cite web|url=https://cctop.ttk.hu/|title=CCTOP|website=cctop.ttk.hu}} HTP database{{cite journal |last1=Dobson |first1=László |last2=Reményi |first2=István |last3=Tusnády |first3=Gábor E. |title=The human transmembrane proteome |journal=Biology Direct |date=28 May 2015 |volume=10 |issue=1 |pages=31 |doi=10.1186/s13062-015-0061-x |pmid=26018427 |pmc=4445273 |doi-access=free }}[http://htp.enzim.hu/ The human transmembrane proteome database] provides a collection of topologies that are computationally predicted for human transmembrane proteins.

Discrimination of signal peptides and transmembrane segments is an additional problem in topology prediction treated with a limited success by different methods.{{cite journal |last1=E. Tusnady |first1=Gabor |last2=Simon |first2=Istvan |s2cid=6431228 |title=Topology Prediction of Helical Transmembrane Proteins: How Far Have We Reached? |journal=Current Protein & Peptide Science |date=1 November 2010 |volume=11 |issue=7 |pages=550–561 |doi=10.2174/138920310794109184 |pmid=20887261 |url=http://pdfs.semanticscholar.org/cf79/b5df3a775212fecda3aeee57104de8f6c382.pdf |archive-url=https://web.archive.org/web/20190307014958/http://pdfs.semanticscholar.org/cf79/b5df3a775212fecda3aeee57104de8f6c382.pdf |url-status=dead |archive-date=7 March 2019 }} Both signal peptides and transmembrane segments contain hydrophobic regions which form α-helices. This causes the cross-prediction between them, which is a weakness of many transmembrane topology predictors. By predicting signal peptides and transmembrane helices simultaneously (Phobius), the errors caused by cross-prediction are reduced and the performance is substantially increased. Another feature used to increase the accuracy of the prediction is the homology (PolyPhobius).”

It is also possible to predict beta-barrel membrane proteins' topology.{{cite journal |last1=Tsirigos |first1=Konstantinos D. |last2=Elofsson |first2=Arne |last3=Bagos |first3=Pantelis G. |title=PRED-TMBB2: improved topology prediction and detection of beta-barrel outer membrane proteins |journal=Bioinformatics |date=1 September 2016 |volume=32 |issue=17 |pages=i665–i671 |doi=10.1093/bioinformatics/btw444 |pmid=27587687 |doi-access=free }}{{cite journal |last1=Savojardo |first1=Castrense |last2=Fariselli |first2=Piero |last3=Casadio |first3=Rita |title=BETAWARE: a machine-learning tool to detect and predict transmembrane beta-barrel proteins in prokaryotes |journal=Bioinformatics |date=15 February 2013 |volume=29 |issue=4 |pages=504–505 |doi=10.1093/bioinformatics/bts728 |pmid=23297037 |doi-access=free }}