Technology

BIOINFORMATICS

Cloning

Expression

Purification

Crystallisation

Structure

Bioinformatics provides essential information for all phases of Structural Proteomics and is especially useful for Target Selection and Target Optimisation. The SSPF bioinformatics core is located in the Barton group at the University of Dundee. As a consequence, work on SSPF bioinformatics benefits from synergy with active bioinformatics research projects in a range of areas from visualisation (e.g. Jalview), through analysis and prediction of protein-protein interactions, to high-throughput sequence and proteomics analysis.

As well as exploiting a range of publicly available bioinformatics tools (Appendix below), a number of new computational tools and techniques have been developed within SSPF. TarO analyses a protein sequence by a large number of bioinformatics techniques. These include crystallisation propensity prediction, searching, and many other sequence-based calculations. Results are tabulated and available via an annotated multiple sequence alignment, that can be edited interactively in Jalview. The principal focus is to identify sequences that are most likely to yield the molecular structure relevant to a given user-defined function. A further aim is to inform the subsequent phases of the process of obtaining the protein's molecular structure, from cloning to molecular modelling. TarO is still in development; however significant functionality is currently available in TarO version 2.0.

TarO connects to available DAS (distributed annotation system) information via Jalview and links to Dasty2, as well as providing routes to other gateways such as Uniprot, COG and the Conserved Domains Database. More information on TarO is available here and the result of an example TarO query can be browsed here.

Bioinformatics tools for the prediction of protein crystallisation propensity have also been developed within the SSPF. These include the OB-Score, ParCrys and XANNPred.

ParCrys is a Parzen Window approach based on calculated isoelectric point, hydrophobicity and the frequencies of S, C, G, F, Y, M residues. Independent test data finds ParCrys to outperform other publicly available methods. More details on ParCrys are available here. The OB-Score is a Z-score scale based on calculated isoelectric point and hydrophobicity. For more details, see Overton & Barton (2006). FEBS Lett.580, 4005-4009. XANNPred is an Artificial Neural Network approach that is still under development.

Please direct any queries to :
Ian Overton (taro "at" compbio.dundee.ac.uk) or Lester Carter (lc71 "at" st-andrews.ac.uk)

Appendix
Jalview (Clamp et al. (2004) Bioinformatics 20:426)
BLAST, RPSBLAST & PSIBLAST (Altschul et al.(1997) NAR 25:3389)
MUSCLE (Edgar, R. (2004) NAR 32:1792)
Globplot (Linding et al.(2003) NAR 31:3701)
Disembl (Linding et al.(2003) Structure 11:1453)
RONN (Yang et al. (2005) Bioinformatics 21:3369)
SignalP (Emmanauelsson et al. (2007) Nature Protoctols 2:953)
TMHMM2 (Krogh et al. (2001) JMB 305:567)
NetNGlyc (Gupta et al. (2004) http://www.cbs.dtu.dk/services/NetNGlyc/ )
NetOGlyc (Julenius et al.(2005) Glycobiology 15:153)
NetPhos (Blom et al. (1999)JMB 294:1351)
Jpred (Cuff et al. (1998) Bioinformatics 14:892)
COG database (Tatsuov et al. (2003) BMC Bioinformatics 4:41)
PDB (Berman et al. (2007) NAR 35:D301)
TargetDB (Chen et al. (2004) Bioinformatics 20:2680)
InterPro (Apweiler et al.(2001) NAR 29:37-40)
Conserved domain database (CDD) (Marchler-Bauer et al. (2005) NAR 33:D192)
Pfam (Bateman et al. (2004) NAR 32:D138)
SMART (Schultz et al. (1998) PNAS 95:5857)
UniProt (Apweiler et al.(2004) NAR 32:D115)
Bioperl (Stajich et al.(2002) Genome Res. 12:1611)
EMBOSS (Rice et al. (2000) Trends in Genetics 16:276)