Data - deephelicon

Input¶

External tools¶

Protein sequences are only needed for running DeepHelicon, and are also used to generate a list of intermediate files before running DeepHelicon, as shown in Table 1.

Table 1:External tools for generating intermediate files before running DeepHelicon.

Tool	Role	Function	Source
HHblits	Input	generating multiple sequence alignments	Remmert et al. (2011)
CCMPred	Input	predictor of residue contacts	Seemayer et al. (2014)
plmDCA	Input	predictor of residue contacts	Ekeberg et al. (2013)
Gaussian DCA	Input	predictor of residue contacts	Baldassi et al. (2014)
Freecontact	Input	predictor of residue contacts	Kaján et al. (2014)
TMHMM	Input	predictor of transmembrane topologies	Krogh et al. (2001)
Uniclust30 database	Intermediate	sequence database	Mirdita et al. (2016)

Output files¶

DeepHelicon can return an output file with the suffix of .s1 (stage 1), .s2i1 (stage 2), .s2i2 (stage 2), .s2i3 (stage 2), .s2i4 (stage 2) or .deephelicon depending on different models used at different stages.

This predictor returns predictions of inter-helical residue contacts in tansmembrane proteins. If non-transmembrane segment or <1 transmembrane segment is detected, the programe will not return final results. However, you can still utilise the intermediate results at stage 1 and 2 as stated in the paper Sun & Frishman (2020). Considering <1 helix detection by inside transmembrane topology predictor, we will consider extending our module to generate a file including entire results in the future work. DeepHelicon outputs results in two formats.

Deephelicon-format¶

Prediction results of interaction sites in tansmembrane proteins consist of 5 columns.

Table 2:DeepHelicon output format.

Position 1	Residue 1	Position 2	Residue 2	Probability
1	S	6	L	0.14790976
1	S	7	R	0.041100707
1	S	8	W	0.04841847
...	...	...	...	...
170	F	176	K	0.05994133
171	A	176	K	0.07471807

CASP14 format¶

CASP14 output has 3 columns: positions of residue pairs and their contact probabilities.

Table 3:DeepHelicon output format.

Position 1	Position 2	Probability
1	6	.148
1	7	.041
1	8	.048
...	...	...
170	176	.060
171	176	.075

Example data¶

Users can download some example data and check an assortment of input files.

Code

Output

import deephelicon

deephelicon.predict.download_data(
    url='https://github.com/2003100127/deephelicon/releases/download/example_data/example_data.zip',
    sv_fpn='../../data/deephelicon/example_data.zip',
)

 ____                  _   _      _ _                 
|  _ \  ___  ___ _ __ | | | | ___| (_) ___ ___  _ __  
| | | |/ _ \/ _ \ '_ \| |_| |/ _ \ | |/ __/ _ \| '_ \ 
| |_| |  __/  __/ |_) |  _  |  __/ | | (_| (_) | | | |
|____/ \___|\___| .__/|_| |_|\___|_|_|\___\___/|_| |_|
                |_|                                   

06/04/2025 04:45:18 logger: =>Downloading starts...
06/04/2025 04:45:20 logger: =>downloaded.

References¶

Remmert, M., Biegert, A., Hauser, A., & Söding, J. (2011). HHblits: lightning-fast iterative protein sequence searching by HMM-HMM alignment. Nature Methods, 9(2), 173–175. 10.1038/nmeth.1818
Seemayer, S., Gruber, M., & Söding, J. (2014). CCMpred—fast and precise prediction of protein residue–residue contacts from correlated mutations. Bioinformatics, 30(21), 3128–3130. 10.1093/bioinformatics/btu500
Ekeberg, M., Lövkvist, C., Lan, Y., Weigt, M., & Aurell, E. (2013). Improved contact prediction in proteins: Using pseudolikelihoods to infer Potts models. Physical Review E, 87(1). 10.1103/physreve.87.012707
Baldassi, C., Zamparo, M., Feinauer, C., Procaccini, A., Zecchina, R., Weigt, M., & Pagnani, A. (2014). Fast and Accurate Multivariate Gaussian Modeling of Protein Families: Predicting Residue Contacts and Protein-Interaction Partners. PLoS ONE, 9(3), e92721. 10.1371/journal.pone.0092721
Kaján, L., Hopf, T. A., Kalaš, M., Marks, D. S., & Rost, B. (2014). FreeContact: fast and free software for protein contact prediction from residue co-evolution. BMC Bioinformatics, 15(1). 10.1186/1471-2105-15-85
Krogh, A., Larsson, B., von Heijne, G., & Sonnhammer, E. L. L. (2001). Predicting transmembrane protein topology with a hidden markov model: application to complete genomes11Edited by F. Cohen. Journal of Molecular Biology, 305(3), 567–580. 10.1006/jmbi.2000.4315
Mirdita, M., von den Driesch, L., Galiez, C., Martin, M. J., Söding, J., & Steinegger, M. (2016). Uniclust databases of clustered and deeply annotated protein sequences and alignments. Nucleic Acids Research, 45(D1), D170–D176. 10.1093/nar/gkw1081
Marks, D. S., Colwell, L. J., Sheridan, R., Hopf, T. A., Pagnani, A., Zecchina, R., & Sander, C. (2011). Protein 3D Structure Computed from Evolutionary Sequence Variation. PLoS ONE, 6(12), e28766. 10.1371/journal.pone.0028766
Sun, J., & Frishman, D. (2020). DeepHelicon: Accurate prediction of inter-helical residue contacts in transmembrane proteins by residual neural networks. Journal of Structural Biology, 212(1), 107574. https://doi.org/10.1016/j.jsb.2020.107574