Network Protein Sequence

Figure 12.6. Secondary structure prediction of duck lysozyme at NPS . The predicted secondary structures of duck lysozyme at Network Protein Sequence Analysis (NPS ) with GOR IV method are depicted in different representations.

Figure 12,7. Secondary structure consensus prediction at NPS . Network Protein Sequence Analysis (NPS ) offers numerous methods for secondary structure prediction of proteins. The secondary structure consensus prediction (Sec.Cons.) of duck lysozyme is derived from simultaneous execution of predictions with more than one methods.

The statistical methods for predicting secondary structures of proteins from amino acid sequences are widely practiced among investigators in biochemistry and can be accessed at Network Protein Sequence Analysis (NPS ) via http //npsa-pbil.ibcp.fr... 

It has also been demonstrated that the neural network approaeh ean be utihsed to predict the 3-D backbone folding of a protein related to the proteins the neural network has been trained upon (Bohr et al, 1990). A neural network is trained upon matehing pairs of protein sequences and secondary stractrrre as well as C a distance eorrstraints (i.e. the distance between 2 residue Ca-atoms larger or smaller than a seleeted threshold... 

The Sequence Retrieval System (Etzold et ah, 1996) is a network browser for databases at EBI. The system allows users to retrieve, link, and access entries from all the interconnected resources such as nucleic acid, EST, protein sequence, protein pattern, protein structure, specialist/boutique, and/or bibliographic databases. The SRS is also a database browser of DDBJ, ExPASy, and a number of servers as the query system. The SRS can be accessed from EBI Tools server at http // www2.ebi.ac.uk/Tools/index.html or directly at http //srs6.ebi.ac.uk/. The SRS permits users to formulate queries across a range of different database types via a single interface in three different methods (Figure 3.4) ... 

The method of Rost and Sander (Rost and Sander, 1993), which combines neural networks with multiple sequence alignments known as PHD, is available from the PredictProtein (Rost, 1996) server of Columbia University (http //cubic.bioc. columbia.edu/predictprotein/). This Web site offers the comprehensive protein sequence analysis and structure prediction (Figure 12.8). For the secondary structure prediction, choose Submit a protein sequence for prediction to open the submission form. Enter e-mail address, paste the sequence, choose options, and then click the... 

Although the existence of neutral networks has been well established for RNA, it is important to note the difference between RNA and protein landscapes. Most important, all RNA sequences fold into some structure, whereas it is likely that the vast majority of protein sequence space is devoid of well-defined structure. It may be possible to overcome this problem because protein sequence space has a higher dimensionality, there are more chances to produce a connected network. In a preliminary study, Stadler and coworkers estimated the neutral structure of... 

Wu, C. H Berry, M., Shivakumar, S. McLarty, J. (1995). Neural networks for full-scale protein sequence classification sequence encoding with singular value decomposition. Machine Learning 21,177-93. 

An interesting example of using Kohonen maps for the analysis of protein sequences is given in a journal article by Hanke (Hanke Reich, 1996). In this application, a trained Kohonen map network was used to identify protein families, aligned sequences or segments of similar secondary structure, in a highly visual manner. 

Ferran, E. A. Pflugfelder, B. (1993). A hybrid method to cluster protein sequences based on statistics and artificial neural networks. ComputAppl Biosci 9,671-80. 

Milik et al. (1995) developed a neural network system to evaluate side-chain packing in protein structures. Instead of using protein sequence as input to the neural network as in most other studies, protein structure represented by a side-chain-side-chain contact map was used. Contact maps of globular protein structures in the Protein Data Bank were scanned using 7x7 windows, and converted to 49 binary numbers for the neural network input. One output unit was used to determine whether the contact pattern is popular in... 

Table 11.1 Neural network applications for protein sequence analysis.

Neural network applications for protein sequence analysis are summarized in Table 11.1. Like the DNA coding region recognition problem, signal peptide prediction (11.2) involves both search for content and search for signal tasks. An effective means for protein sequence analysis is reverse database searching to detect functional motifs or sites (11.3) and identify protein families (11.4). Most of the functional motifs are also... 

Neural networks have been used successfully for the detection of binding or regulatory sites in nucleic acid sequences that lack clear consensus sequences. This is also true for the prediction of cleavage or acceptor sites in protein sequences. 

There are two different approaches to the protein sequence classification problem. One can use an unsupervised neural network to group proteins if there is no knowledge of the number and composition of final clusters (e.g., Ferran Ferrara, 1992). Or one can use supervised networks to classify sequences into known (existing) protein families (e.g., Wu et al., 1992). 

Wu et al. (1992) devised a neural network system for the automatic classification of protein sequences according to superfamilies. It was extended into a full-scale system for classification of more than 3,300 PIR protein superfamilies (Wu et al., 1995). The basic input information was encoded with the n-gram and SVD methods. The implementation... 

As a full-scale family classification system, more than 1200 MOTIFIND neural networks were implemented, one for each ProSite protein group. The training set for the neural networks consisted of both positive (ProSite family members) and negative (randomly selected non-members) sequences at a ratio of 1 to 2. ProClass groups non-redundant SwissProt and PIR protein sequence entries into families as defined collectively by PIR superfamilies and ProSite patterns. By joining global and motif similarities in a single classification scheme, ProClass helps to reveal domain and family relationships, and classify multi-domained proteins. 

How have neural networks been used in genome informatics applications In Part II, we have summarized them based on the types of applications for DNA sequence analysis, protein structure prediction and protein sequence analysis. Indeed, the development of neural network applications over the years has resulted in many successful and widely used systems. Current state-of-the-art systems include those for gene recognition, secondary structure prediction, protein classification, signal peptide recognition, and peptide design, to name just a few. 

---