Molecular Biology (Soaring Bear)

home	SOARING BEAR Ph.D. Pharmacology
index	Scientific advisor to the health professions & herbal industry	index

Bioinformatics, Sequence and Analysis - genomics & proteomics

Bioinformatics Courses: NCBI GeneAnal - NCBI - geneAnal - geneAnal2006 - geneAnal2003 - S-Star - Altman Stanford - Attwood UCL - Attwood UCL - Bourne SDSC - Church Harvard - Condon UBritCol - Condon UBC - DeonierWaterman USC - Fristensky UManitoba - Gerstein Yale - Soll Yale - Guttman UToronto - Houry UToronto - Hunter UColo - Hunter UColo - UMass - Momand CSULA - Momand CSULA - Mount UAriz - Nirav UAriz - Murphy CMU - Ruzzo UWash - Skiena SUNY SB - Skiena SUNYSB - Shamir TelAvivU - Shamir TelAvivU - Shamir TelAvivU - Stein TelAvivU - Ruzzo Tompa UWash - Weng BU - Weng BostonU - Reimers Karolinska - Reimers Karolinska - Karolinska -CBS - SwissEMB - Swiss Inst Bioinf - UK HGMP - VSNS - webbooks - cybertory - geneexploration(nature) - star - NIHgene analysis course - Max Planck course - ergito (sbear@uswest) - GeneticsHomeReference(NLM) - VNPS -

Bioinformatics business news: beye -

biotech directory

Chemical Structure page may be of interest

Chemogenomics: Iconix -

multiple db links: dbget - embl -

GeneLiteratureClustering(pubgene) - iHOP geneLit relationship -

SEQUENCE: genebank

Comprehensive SEQUENCE PATTERNS
Tutorials: Geof Barton - domain algorithms - Seq Anal Intro & Links - CyberConnect -

BLAST (Basic Local Alignment Search Tool) is the most widely used method to find other sequences similar to one you have. It is optimized for speed so if a more rigorous method is desired use Smith Waterman (see comprehensive below). Many sites are available on the web but results differ because of different versions of the program, different parameters that can be set and different databases searched against. Defaults work well with medium length sequences (nucleotides 100-1000, proteins 50-200).
Protein sequence is better to search on than DNA because: (a) information content per residue is greater (20 letters vs. 4); (b) nonidentical aa can be scored by mutation frequency matrix like PAM or BLOSOM; (c) protein database is smaller so chance matches are reduced.

	sequence database searched	sequence in hand	use
blastn	nucleotide	nucleotide	check if a newly obtained sequence is already published identify coding regions by searching for mRNA sequences extend sequence fragments by in EST db
blastp	protein	protein	identify new proteins, find homologs, identify domain structure by related proteins of known structure
blastx	protein	nucleotide translated	identify potential coding regions in newly sequenced DNA
tblastn	nucleotide translated	protein	detect unidentified proteins in a nucleotide database
tblastx	nucleotide translated	nucleotide translated	increased sensitivity but slower than direct blastn

service	_p	_n	_x	t_n	t_x	fasta	SW	notes
EBI	*	*	*
Sanger		*		*	*
NCBI	*	*	*	*	* email			graphical overview
BioNavigator$	*	*	*	*	*		*	results easily saved
GWFASTA						*		imtech
Raghava						*		imtech
TIGR

    If you have a long sequence and are hunting for short domains, then splitting into overlapping fragments is best.
    Searching with a short sequence often gets nothing because BLAST filters out random similarity. For PCR primers increase the statistical expectation (expect in NCBI, maxexp in EBI).
    Repeats & regions of low complexity are usually filtered out to avoid false positives. Masked region appears as a string of N or X so it may be useful to repeat with this turned off.
    Distant homology can be sought by adjusting scoring matrix. The usual BLOSUM62 may be lowered to 45 or the usual PAM30 increased to 250. BLOSUM (derived from BLOCKS protein conserved regions) is considered better than the older PAM (derived from a smaller set of globular proteins).
    BLAST Documentation: Blast Intro - Blast Manual - Blast Ex output - Blast Help - Gilbert blast script - Matrices

Pairwise Alignment: BLASTULA links - BLAST (nlm) - ALIGN, LFASTA, SIM (BCM) - ALIGN (Genestream) - SIM (expasy) - SIM, GAP, GAP2, LAP2 (Mich Tech) - ALIGN, Wise2 (pasteur) -

PATTERNS of Protein/DNA sequences

BLAST and FASTA:
BCM (also MSA, patterns, YAC) - GenQuest - ebi blast against pdb - GeneStream -

FASTA (Pearson, UVa) more sensitive & slower than blast: ebi - ebi - eerie - eerie - ORNL - heidelberg prosite - Munich IPS - IBCP Lyon - pasteur - jhu genquest - tsukuba - ncc japan mailback - singapore mailback - Docs: Intro or algorithm - example - Pearson on ver2 Software: download

BLAST (compare sequence to db) faster: ncbi-blast - ncbi - cam.ac via ncbi - HIV(ncbi) - Immunoglobulin Blast(ncbi) - Prodom-toulouse - Prodom-sanger - Prodom doc - trieste sbase - SCOP - CSC - tsukuba - tsukuba links -

OTHER Patterns: BLASTPAT - FASTPAT - patternhunter - Sequerome -
BNL (MSA, stru, pI, mw, Blast, homol, blast, genbank, primer, restric sites, etc)
- bmerc ProLink homol - bmerc prot seq anal - Darwin (cbrg) AllAll - BioSCAN match seq or name to db

Smith Waterman algorith is more rigorous than Blast: EMBnet comparison
SW servers: GeneStream - SAMBA - embnet - EBI - stanford -

EBI: fasta, blitz, prosite, blast, bcm results emailed

curatools - bioperl scripts -

exon-exon splice prediction(BCM) - ExonIntronDb(Gilbert Harvard) -

HT-SAS annotation service - HTSAS -

Bioinformatics free:
Galaxy - Galaxy - Galaxydoc -
Bioweb -
Magallanes - magallanes -
Curatools: Genescape, blast, clustal, primer, ORF, RE, prosite, blocks, prodom, PHYLIP, etc
NCI cancer gene anatomy project - NCI bioinformatics -
NCSA workbench: MSA, motif, protein structure prediction, etc
UK Human Genome Mapping: GCG, Staden, GDE, PHYLIP, etc
Compugen LabonWeb (4k human genes, 10k sequences, Genzyme's Sage)
PubGene
Lexicon Lexgen Omnibank: mice & human
Stothard links to sequence analysis tools and Jscripts - Stothard/UAlberta sequence manipulation -
Manchester
vectorNTI: sequence analysis, alignment, primer design
genemine
?Doubletwist: (Pangea) EST consensus, ProNet protein interactions
wikigenes -

Integrated Bioinformatics Search:
BCM:
SearchPalette (Wuerzburg)
Weizmann & Skoufos

Bioinformatics fee$:
GCG/Vaiken
ANGIS & eBioinformatics (Australia, BioNavigator) GCG, etc
Bionavigator: clustal, motifs, genbank, fasta, swisprot, primer, etc)
Hyseq GeneSolutions: 12 million EST/DNA seq
Canadian: GCG, etc
Bioinf Inst India
gened $300

Comprehensive SEQUENCE ALIGNMENT

MSA (Multiple Sequence Allignment): WUSTL (sw/pir#) - BCM pairs - Fullen index - Clustal, PIMA, MSA, MAP (bcm) - ToPLign (Sankt) - Karolinska - MAP (Mich Tech) - Pimall (BU) - MultAlign (Zurich) - AllAll (Zurich) - MPSA - SAM (UCSC - viseur (nancy) - Musequal2 (nih) - cbrg AllAll (#;,#;) - Gryan's FALCON-large contigs - proanwin (ebi) - proanwin - map(C) - VSNS Bielefeld list -
MSA Clustal, progressive alignment starts with most similar portions: Clustal(bcm) - Clustal (ebi) - Clustal (emory bimcore) - Clustal (nih) - Clustal (pasteur) - Clsutal(Lyonnais) - Clustal(jp) - Clustal(Transfac) - Bionavigator - Clustal tutorial - Clustal tutorial -
MSA Pileup, progressive alignment using UPGMA: GCG - UA - NIH - HGMP - Bionavigator -
MSA Dialign, local alignment: dialign (pasteur) - dialign(Genomatix) - dialign(Bielefeld) -
Documentation: algorithms - conserved blocks algorithm - alscript seq->ps - ShadyBox (emory bimcore) -

Alignments Analysis: AMAS alignment - WebLogo - Hidden Markov Model - SAM HMM - Profile-SS (Watterman-Eggert) - SAM (Haussler UCSC) - Smith Waterman (paracel) - emboss - monash - monash - Fuellen basics - Fuellen gentle guide - Fuellen gentle-bcm - Fuellen gentle-uk - Tekaia MSA - bcm - vsns - vsns (bcm) - vsns (it) - barton - barton mirror - bcd bielefeld - bcd bcm - bcd uk - gibbs -

Multiple Alignments Databases: ALI search - ALI info FSSP folds - ShadyBox - BoxShade -

Temperature Melt (Tm) prediction: MeltDNA -

protein structure alignment servers and programs: CE (download) uses extension of the optimal path - DALI (download) uses distance matrix - KENOBI uses genetic algorithm - PRISM (download) uses sse alignment plus iterative refinement - PROSUP (download) uses hierarchical alignment (to build initial equivalence list) followed by dynamic programming refinement - SAP (download) uses double dynamic programming - SHEBA (download) uses hierarchical alignment with profiles - TOP (download) uses sse alignment - VAST uses vector alignment - STRUCTAL (download) uses double dynamic programming -

Database of Genome Sizes (DOGS)
PubDNAfinder -

ANALYSIS of Protein Sequences

Function from sequence: uniprot - prophyler - aranet - seq2ref -

BCM Beauty Protein blast, prosite, blocks - predictprotein - Attwood UCLondon -
Motifs: Bork/sander motifs - Promotif, threader, etc (bsm) - motifs - nuclear localiz motif - Eisenberg - Eisenberg - Eisenberg article - Zincbinding - smart - pfam - pfam - deepview -
BLOCKS: protein motifs/homol (Hutchinson): Blocks - Blocks help gopher amino acids -
PROSITE: protein motifs (Blocks has more) (expasy, assoc with Swiss-Prot): expasy prosite allows direct blast - heidelberg prosite - harvard search - table of motifs-harvard - Japan Prosite/MotifDic - nih search - GCG profile-weizmann - UI search . Docs: UI isrec Prosite Guide

Prints motifs & domains - GTlinks - DomainFinder -
EMBL interpro - interpro -

Prot Alignment (Barton) - Matchbox MSA protein blocks, no gap penalty: Matchbox to 2000 aa (>nameCRseq) - BioSCAN match seq or ac# to db (blosum/pam, MSA) -

cbrg all/all match Swiss-Prot mailback

PDBTool-structure validation

pro-fit - ProFit

fpat - Munich - ProClass (Wu) - xtal packing - Labvelocity links for seq analysis -
PSORT (prediction of protein sorting signals and localization)
Domains: Domains db - DomainFinder -
protein folding concepts: klimov -
amino acid side chain interactions: Singh & Thornton - Ramachandran -
ProtoMap - ProtoMap (clustering of Swiss-Prot proteins; Yona, Proteins 37(3)360-78 1999)
Signal Transduction: STKE(Science, sbear j 10859764) -
ProFound peptide mapping
Profiles-3D for protein model verification (Eisenberg): ucla - ucla - msi -

Genome-based Peptide Fingerprint Scanning (unk prot searched against genes)

mowse protein MW calc
Phosphorylation sites (Ser,Thr,Tyr): phosphoELM -

SequenceMusic: whozoo - nslij - ucla - Takahashi -

SEQUENCE SERVERS

There are significant differences in how well defined the results are and in how succintly the information is displayed. Some display every little fragment ever reported plus extraneous unrelated ones too, while others just show complete sequences. There are variations in availability of boolean operators to narrow - widen searches. Some hyperlink further on to medline reference text and motif servers. Some require more steps of linking (and waiting for server response) than others to get the desired information. It is unfortunate that at the present time many of the servers are not able to distinguish in the search between topoisomerase I and topoisomerase II and separating them is still a manual chore. An evaluation chart might look like:
database choices:
1 line - complex form
Example number of hits on:
topoisomerase
EC# 5.99.1.?
topoisomerase I
EC# 5.99.1.2
topoisomerase II
EC# 5.99.1.3
Boolean operators
Provides on First screen: AC#, code (name, species), EC#
## links to see seq
seq formats: fasta, ncbi
Links to: prosite, entrez, medline, prodom, neighbors, gdb/embl
able to refer to address later?
comments
notes: n-narrow ( only complete sequences - less than expected number ) ; w-wide ( every little redundant fragment ) ; x-extraneous unrelated hits .

Comprehensive SRS SEQUENCE SEARCH

Sequence Retrieval System (SRS) servers: lionbio - ebi - embl - sanger - .
All have seq lib (pir, swissprot, gene) ; formats: (pir-all, gcg, fasta) ; patterns (prosite, + ) . Provide sw code - pir ac#. Link to complete entries (can keep) - to seq & on to prosite, medline/entrez, genbank/embl, pir/swissprot. Boolean: |=or, &=and, !=but not. Docs .

bio oslo
prot stru, patrn (11), ent-medline, other-limb, transfac
topo&II (51), topo (167) 80sw, 87 pir; link to entrez.

IU
           patrn (8),           other-limb
topoII (87x); topo (233x) 141sw, 92pir; link to entrez

embl heidelberg
prot stru, patrn (15), medlars, other-limb, pdbfind, transfac
topo&II (49); topo (163), 78sw, 85pir; link to medlars

csc finland
prot stru, patrn (10),         other-limb, pdbfind, transfac, taxon
topo&II (51), topo (167) 80sw, 87 pir; link to entrez.

ebi - Ebi
prot stru, patrn (12), medlars, other-limb,          transfac, tags
topo&II (52), topo (167) 80sw, 86pir; link to medlars

weizmann
           patrn (5),                                transfac
topo&II (48), topo (179) 95sw, 84pir; medlars

caos nijmejen
           patrn (15), medlars, other-limb nakai, transfac, lista
topo&II (35); topo (161) 80sw, 81pir; link to medlars

inserm france
           patrn (15),           other-limb, gene, transfac, lista
topo&II (32), topo (157) 76sw, 81pir; link to entrez.

sanger uk
prot stru, patrn ?                                            lista
topo&II (50), topo (166) 79sw, 87pir; link to entrez.

abc hungary
           patrn (3),           other-limb .
topo&II (45), topo (151) 66sw, 85pir; link to entrez.

bioz basel embnet
           patrn (15),           gene lista,         transfac, lista
topo&II (47), topo (159) 72sw (no new), 87pir; link to entrez

bmc uppsala
           patrn (9),           other-limb, transfac
topo&II (42), topo (138) 66sw (no new), 72pir; link to medlars

ben
           patrn (15),          other-limb
topo&II (32), topo (111) 80sw, 31pir; link to entrez.

other menus:
Entrez (ncbi) stru - Entrez (ncbi) - Provides unique name, 200 limit. Links to sequence report, fasta format, medline, neighbors, genbank. topo&II (25), topo (218)
bimas Genobase/nih
bcm BioControl Panel (Genbank, GDB, GSDB, Genethon, Swiss-Prot, PIR, Blast, etc) - FAQ
bioCUSI (Genbank, Sw, Entrez medline 200)
GDB Gopher Other Genbank, Swiss-Prot, PIR, PDB, EMBL, Prosite, Blocks, Patents, Rebase etc)
CSC (Genbank, Swiss-Prot, PIR, PDB, EMBL, Entrez)
NIH (Genbank, Swiss-Prot, PIR, Prosite, Transcript Factors)
NIH GenoBase (Genbank, EMBL, SwisProt, EC, Prosite?, Selkov Enzym-Metab Path
harvard (Genbank, Swiss-Prot, PIR, GCG doc, Medline, etc)
IU (Genbank - PIR or Prosite or Swiss-Prot)
BioSCAN by AC# - codename (Genbank, Swiss-Prot, PIR)
also match seq or name to dbases.
IncyteGenblast

ACNUC seq anal

attotron seq anal

NARlinks(2005) - NARlinks(1/01) - NARlinks(1/01) -

Predicting Protein STRUCTURE from sequence
3DCrunch (200k seq vs PDB) - Robetta (David Baker UWa) - Quadratic Logistic (NIH) - CMS tools - PPS2 - descfold - UCLA-DOE Fold Recognition irbm prot stru prediction - embl - ProFit superimpose - PsiPred - CBS - GPCR Viseur - GPCR expasy- GPCRdb - trans membrane (embl) - accelrys polymorph - upack - trans membrane (unil)?-

neuronet methods: kcl - neuralweb - cmu - embl/sander - embl/heid - Wu - ucsf - Strelets - Schulze-Kremer Genetic Algorithms -

BCM Secondary Stru - Rost MaxHomol (embl) - cpe-phd - Protein db-Elba - Zvelebil 2ndary pred - SCRI 2ndary pred - BSM slo - Naomi - ftp naomi - cornell protein folding doc - UCLA fold recognition - stanford protein domain motion set - stanford protein domain motion set - GBarton Sequence - Expasy: pI, MW, seq anal links - Martin's protein loops - side chain rotamers - Swissmodl (glaxo) & homology - bsm CATH prot groups - Thornton UCL - Overington motifs - Miller predict - XABgen antibody homology - Martin's antibody modeling -

BETAWRAP prediction of parallel B-helices (more common in toxins & disease surface proteins): MIT - PNASarticle -

Homology requires ~25% sequence identity (Structure 4:1123 1996) - casp : Swissmodl (glaxo) & homology - SCWRL - ExpasyDeepview - Modeller- Modeller - Modeller - MolIDE - XABgen antibody homology - bmerc ProLink homol - Rost MaxHomol (embl) - SegMod - SegMod validation - Look - Jackal - populus - plop - phyre - mgenThreader - sam - ffaso - esypred3d - 3dJigsaw - loopp - tome - genesilico - 3dJury - wiki -
Loops: Martin's protein loops - side chain rotamers - loops - modeller - ModLoop - VriendWhatif - SCWRL - Tasser -

BLAST search of 929 human disease genes against Drosophila found 548 (representing 714 diseases) with high similarity in predicted amino acid sequences [Bier, Genome Res 2001 Jun;11(6):1114-25; PMID: 11381037] sdsc - ucsd - abst

homology of organisms (IU) - human mouse homology

Glycosylation prediction - GlycoNet -

sesam struct groups

fssp prot fold families

prot/dna sequence analysis: - BNL - curatools -

memsat-hydrophobicity -

exptl membrane prediction

prot fold id - prot fold id ftp

Procheck

SOPMA prediction

Proteomics software: NIH -
Proteomics tools: Scientist 10/01 - expasy links - expasy links

prot stru conferences -

Structural Motifs; Threading:
Promotif, threader, etc (bsm) dali - PROSPECT - mailPROSPECTlist - joinPROSPECTlist (subscribe prospect-digest) - CN3D - raptorThreading -

PHYLOGENY: embl/SHOT - treejuxtaposer - OneZoom - OneZoom - ETH - ua phylogeny - EBI - EBI - HennigCladistics - Treeview - beast - pasteur - pasteur - TreeGen (cbrg darwin) - UATreeLifeTaxonEvol - UA Tree of Life - TreeofLife - UA - phylo software rev - berkeley - berkeleyTree - gene linkage - ArboDraw - jklustor -
PHYLIP: pasteur doc - bimcore doc - UW source
PAUP: sinauer -

Primer design: Marburg - JHU - PathoGene -

Misc Gluco-Amylase ref page protein disease db -

Transcript Factors: nih tfd seq db - ncbi tfd db - CheaDB -

Embnet news: ire - be - no - uk - emb

Readseq Converter

Staden: manual - archive - Guide - summary

SEQNET Prot Anal ? (GCG, Blast, Staden,etc) -

Documentation & Explainations: G Barton - F Tekaia - Karplus - biocomputing course - draft Algorithms course - doelz - UCSF seq service - Programs summary-RU

GCG index/contents: ae / ae , - usf , - usf / usf , - strasb / strasbg , - vjf / vjf , - pasteur / past .
GCG docs: genhelp - genmanual - protein anal (s) - usf user manual - ae docs - agrenet tutorial - GCG summary-RU - GCG graphical front-Emory Bimcore and manual - Wisc GCG Guide - trieste

IUBio Software

PROTEIN Sequence Databases (also use Comprehensive tools)

EBI PICR Protein Cross Reference of 20 db

ENTREZ protein Provides species code name, 200 limit. Links to seq, fasta format, medline, neighbors, genbank.

PIR: prot & gene DB (Georgetown): genbank, Japan, Refbase . Docs: gdb - gdb - bchs .
GDB PIR fields - GDB PIR keyword : Provides pir#, name, species. Can't keep result location. Links to sequence and further to BLAST/FASTA.
nih PIR (search by: term1 and term2) : Provides pir#, name, EC#, species. Links to sequence.
bchs PIR gopher?

SWISS-PROT: EMBL translations, PIR, refs to Prosite & PDB .
Docs: expasy UI - emiliano - ebi - SwissInstBioinfo -
expasy recent - expasy text - expasy ac# Provides code name, EC#, species. Link to seq & on to PROSITE, medline, PRODOM, etc.
GDB Enzyme Provides ac# & species code name. Links to expasy sequence and can link further to prosite, Prodom, genbank, etc.
ui Provides ac#, name, EC#. Link to seq.

OWL (nonredundant Swiss-Prot, PIR) Provides enzyme name (nonunique). Links to header and can link further to sequence, PIR, EMBL, Genbank, Prosite, etc.

BioSCAN retrieve by AC# - code name . Provides fasta, staden or raw complete formats; BLOSUM/pam patterns.

ebi swprot+pir+gb+pdb

prot seq db links