Applications

Applications

Using standard Bioinformatics applications

Introduction

The overall plan for the regeneration of high quality annotation information as

contained in the EMBL disk-file ISTN501

figWHAT.eps

Scientific Background ToMer Operon

● Function

● Genetic Structure and Regulation

● Mobility Of The Mer Operon

The principal proteins and their functions

figPRINCIPLE.eps

Downloading The Raw DNA Sequence

Initial BLAST Sequence Similarity Search

Maxim 18.1

With BLAST scores, up is down and lower is better

http://opal.biology.gatech.edu/GeneMark/

GeneMark

The web-based interface to GeneMark as running at EBI

figEBIGENEBANK.eps

Using BLAST to identify specific sequences

Dealing with false negatives and missing proteins

Over predicted genes and false positives

http://www.expasy.org/swissmod/

Structural Prediction With SWISS-MODEL

Maxim 18.2

The major limitation of ``homology modelling'' is that homology to a known structure is needed

Alternatives to homology modelling

Modelling with SWISS-MODEL

The SWISS-MODEL predicted structure of ORF2/MerP

figORF2MERP.eps

The SWISS-MODEL predicted structure of ORF2/MerP, second version

figORF2MERP2.eps

The SWISS-MODEL predicted structure of ORF3/MerA (A)

figORF3MERAA.eps

The SWISS-MODEL predicted structure of ORF3/MerAB

figORF3MERAB.eps

The SWISS-MODEL predicted structure of ORF6/TNR5

figORF6TNR5.eps

DeepView as a Structural Alignment Tool

The ORF2 and ORF3_A structures loaded into DeepView prior to structural

alignment

figDEEPVIEW.eps

DeepView's Iterative Magic Fit dialogue box

figDEEPVIEWDIALOG.eps

Structural Alignment created using the DeepView's Iterative Magic Fit facility

figDEEPVIEWEXAMPLE.eps

Selecting the current ``layer'' in DeepView

figDEEPLAYER.eps

Possible Explanation Behind MerA/HMA Duplication Event

figPOSSIBLE.eps

The structural alignment of ORF3_B and the ``official'' Mercury Reductase X-ray

structure

figCYSTEINES.eps

Maxim 18.3

Homology modelling can only model protein sequences similar to those which are already

known

PROSITE and Sequence Motifs

Maxim 18.4

Searching large datasets with non-specific, short sequence fragments results in many false

positives

http://www.expasy.org/prosite/

http://www.ebi.ac.uk/interpro/

http://www.geneontology.org

● http://www.kegg.org

Using PROSITE patterns and matrices

Phylogenetics

A look at the HMA domain of MerA and MerP

------------------------------- -------------------------------SWISS-PROT IDs of MerP Proteins SWISS-PROT IDs of MerA Proteins------------------------------- -------------------------------MERP_ACICA MERA_ACICAMERP_ALCSP MERA_ALCSPMERP_PSEAE MERA_BACSRMERP_PSEFL MERA_ENTAGMERP_SALTI MERA_PSEAEMERP_SERMA MERA_PSEFLMERP_SHEPU MERA_SERMAMERP_SHIFL MERA_SHEPUMERA_SHIFLMERA_STAEPMERA_STRLIMERA_THIFE------------------------------- -------------------------------

The multiple sequence alignment of the example proteins

figLISTMERAMERP.eps

The EBI's tree graphical display

figTREE.eps

Maxim 18.5

Whenever you make a statement, call for more research (money)!

Maxim 18.6

Database annotation is hard to do well, so be prepared to update it on a regular basis

Maxim 18.7

Automation can be very helpful when creating annotation, but to achieve the highest quality,

humans are needed to make some value judgments

Maxim 18.8

Conclusions are based on the available data which, in this case, is the database annotation

(which may or may not be current)

Where To From Here?