Molecular Structure and Function Prediction

Molecular structure and function prediction deals with three dimensional structures of proteins. It helps in designing of protein as well. This plays a major role in the field of biochemistry, medical biotechnology and theoretical chemistry for drug designing and novel enzyme designing. Irrespective of genetic environment presence molecular structure and function prediction is feasible with double strands of DNA rather than single stranded sequence prediction. Following are the ways to predict molecular structure and function prediction.

• Intron splice sites and branch points
• Gene finding
• Gene expression level
• DNA blending prediction
• Sequence clustering & cluster topology
• Nucleosome positioning signals
• Promoters recognition
• RNA structure prediction
• Protein structure prediction
• Protein function prediction
• Protein family classification
• Protein degradation
• Other functional sites of DNA and RNA

Intron splices sites and branch points:  

Splice junctions help in prediction of sequence which interrupts the RNA and protein expression. Intron splice is a great challenge for bioinformatics field. Introns of S. cerevisiae are highly conserved. This helps in easy prediction of coding and non coding regions of genomic DNA.

Gene finding:

Computational gene finding includes open reading frames (ORF’s) quantification, splice site interruption, translation prediction, modeling of gene and its assembly. Gene finding is being incorporated in all methods.

Gene expression level:

Strength of promoter signals predicts gene expression level. Gene expression level helps in coding the sequence, utilization of codon and codon adaption index. It is profoundly used in sequence statistics.

DNA blending prediction:

Prediction of DNA blending helps in understanding of transcription and sequence curvature.

Sequence clustering & cluster topology:

Clustering helps in forming sequential groups and distance can be predicted. Hidden Markov models lend a hand to map self organized sequences.

Nucleosome positioning signals:

Eukaryotic DNA location is predicted as DNA gets bound to histones which are present in chromatin.

Promoter recognition:

Transcription initiation predicts gene expression when RNA production is catalyzed. It is complex process if distance of DNA signals as substrate for polymerase recognition.

RNA structure prediction:

Secondary structure of transfer RNA, messenger RNA and ribosomal RNA can be easily computed depending on the free energy available for active interaction. It is difficult to predict the loop to loop interactions.

Protein structure prediction:

Machine learning help in prediction of secondary structure, distance between protein residues, protein folding, disulfide linkages, phylogenetic family, topology, major histo compatible complex motifs and transmembrane segments.

Protein function prediction:

Protein design, cleavage sites, localization of peptide bond, glycosylation signals, phosphorylation, post translational modification, signal transduction, protein active site can be determined easily.

Protein family classification:

A neural network helps in prediction of association of protein families, self organizing maps and motif based determination.

Protein degradation:

It is common protein of all living organisms are easily cleaved and recycled. Certain degradative pathways are difficult to understand due to unfolding of protein and specificity.

Other functional sites of DNA and RNA:  

Functions like motif domain in protein DNA interaction, DNA helix category, introns, rRNA classification, ribosomal binding sites, phylogenetic sequencing and classification, tRNA sequence classification and DNA melting points are easily predicted by computational biology.