Exercise: Multiple Alignments (Seaview version) - Revision history

Henni: /* Step 7 */

2024-11-13T14:58:31Z

Step 7

← Older revision		Revision as of 16:58, 13 November 2024
Line 169:		Line 169:
	Align the EPB sequences using Clustal Omega in Seaview, as before. Save it as "<tt>EPB4.1_human.clustalo.fasta</tt>" so you can see by the window title which method you used. Then, open a new window with the same sequences and align them with MUSCLE (set <u>Alignment options</u> to "muscle"). Give that result a recognizable name, too. Compare the two results.		Align the EPB sequences using Clustal Omega in Seaview, as before. Save it as "<tt>EPB4.1_human.clustalo.fasta</tt>" so you can see by the window title which method you used. Then, open a new window with the same sequences and align them with MUSCLE (set <u>Alignment options</u> to "muscle"). Give that result a recognizable name, too. Compare the two results.

	Then, try the MAFFT and Kalign servers at [http://www.ebi.ac.uk/~~Tools~~/msa/ the EBI Multiple Sequence Alignment page] with FASTA as output format (and keep all other options at their default). For each alignment, keep the window (or tab) open after use. Download the aligned sequences and open them in Seaview (''without'' re-aligning them). You should now have four open Seaview windows.		Then, try the MAFFT and Kalign servers at [http://www.ebi.ac.uk/jdispatcher/msa/ the EBI Multiple Sequence Alignment page] with FASTA as output format (and keep all other options at their default). For each alignment, keep the window (or tab) open after use. Download the aligned sequences and open them in Seaview (''without'' re-aligning them). You should now have four open Seaview windows.

	Now compare the the four alignments you just constructed. Remember that an alignment of isoforms ideally should have only matches and gaps, no mismatches. If you find it hard to spot mismatches directly, you can create a 100% consensus sequence as before and search for X'es in that sequence. Ideally, there should not be any.		Now compare the the four alignments you just constructed. Remember that an alignment of isoforms ideally should have only matches and gaps, no mismatches. If you find it hard to spot mismatches directly, you can create a 100% consensus sequence as before and search for X'es in that sequence. Ideally, there should not be any.

Henni: /* Part 1 — using Clustal Omega in Seaview */

2024-11-13T14:57:28Z

Part 1 — using Clustal Omega in Seaview

← Older revision		Revision as of 16:57, 13 November 2024
Line 12:		Line 12:
	[http://www.clustal.org/omega/ Clustal Omega] from 2011 is a successor to the widely used [http://www.clustal.org/clustal2/ ClustalW/ClustalX] package. This came in two varieties: '''ClustalW''' (with a command line interface) and '''ClustalX''' (with a graphical interface). Typically, ClustalX was chosen for interactive use, and ClustalW was chosen when there was a need for automating the workflow. The underlying algorithm was precisely the same, and the results were identical.		[http://www.clustal.org/omega/ Clustal Omega] from 2011 is a successor to the widely used [http://www.clustal.org/clustal2/ ClustalW/ClustalX] package. This came in two varieties: '''ClustalW''' (with a command line interface) and '''ClustalX''' (with a graphical interface). Typically, ClustalX was chosen for interactive use, and ClustalW was chosen when there was a need for automating the workflow. The underlying algorithm was precisely the same, and the results were identical.

	Clustal Omega only comes as a command line interface, but Seaview provides a graphical interface to it. In this exercise (and next week) we will mainly use Seaview to run Clustal Omega. Later in today's exercise, we will also run MUSCLE through Seaview and a couple of other multiple alignment programs via EBI's multiple alignment page: http://www.ebi.ac.uk/~~Tools~~/msa/.		Clustal Omega only comes as a command line interface, but Seaview provides a graphical interface to it. In this exercise (and next week) we will mainly use Seaview to run Clustal Omega. Later in today's exercise, we will also run MUSCLE through Seaview and a couple of other multiple alignment programs via EBI's multiple alignment page: http://www.ebi.ac.uk/jdispatcher/msa/.
	<!--		<!--
	ClustalW/X is free to use and available for many computer systems — including Windows and Mac. If you want to install ClustalX on your own machine, it can be downloaded from here: http://www.clustal.org/clustal2/. It can also be used in an online version under EBI's multiple alignment page: https://www.ebi.ac.uk/jdispatcher/msa.		ClustalW/X is free to use and available for many computer systems — including Windows and Mac. If you want to install ClustalX on your own machine, it can be downloaded from here: http://www.clustal.org/clustal2/. It can also be used in an online version under EBI's multiple alignment page: https://www.ebi.ac.uk/jdispatcher/msa.

Henni: /* Step 7 */

2024-11-13T14:29:58Z

Step 7

← Older revision		Revision as of 16:29, 13 November 2024
Line 165:		Line 165:
	We are now going to investigate how four different alignment programs perform on a dataset of isoforms from one particular gene.		We are now going to investigate how four different alignment programs perform on a dataset of isoforms from one particular gene.

	[~~[Media~~:EPB4.1_human.fasta~~\|'''~~Here''']] is a dataset consisting of 11 alternatively-spliced gene products from the human erythrocyte membrane protein band 4.1 (EPB). The goal of this exercise is to compare how well three different popular multiple alignment programs perform when attempting to align a set of proteins that are identical except for having different deletions.		'''[https://teaching.healthtech.dtu.dk/material/22111/FO2024/EPB4.1_human.fasta Here]''' is a dataset consisting of 11 alternatively-spliced gene products from the human erythrocyte membrane protein band 4.1 (EPB). The goal of this exercise is to compare how well three different popular multiple alignment programs perform when attempting to align a set of proteins that are identical except for having different deletions.

	Align the EPB sequences using Clustal Omega in Seaview, as before. Save it as "<tt>EPB4.1_human.clustalo.fasta</tt>" so you can see by the window title which method you used. Then, open a new window with the same sequences and align them with MUSCLE (set <u>Alignment options</u> to "muscle"). Give that result a recognizable name, too. Compare the two results.		Align the EPB sequences using Clustal Omega in Seaview, as before. Save it as "<tt>EPB4.1_human.clustalo.fasta</tt>" so you can see by the window title which method you used. Then, open a new window with the same sequences and align them with MUSCLE (set <u>Alignment options</u> to "muscle"). Give that result a recognizable name, too. Compare the two results.

Henni: /* Step 8 */

2024-11-11T16:25:38Z

Step 8

← Older revision		Revision as of 18:25, 11 November 2024
Line 201:		Line 201:

	'''Closing remark''': Currently the RevTRans server does not perform a lot of additional analysis on top of the actual alignment. The idea with the server is to provide input to "downstream" analysis in other tools. For example construction of phylogenetic trees and statistical analysis of silent vs. non-silent mutations (that is, mutations that do not change the amino acid sequence vs. mutations that do).		'''Closing remark''': Currently the RevTRans server does not perform a lot of additional analysis on top of the actual alignment. The idea with the server is to provide input to "downstream" analysis in other tools. For example construction of phylogenetic trees and statistical analysis of silent vs. non-silent mutations (that is, mutations that do not change the amino acid sequence vs. mutations that do).

			'''The new closing remark (added 2024)''': Actually, you don't really need the RevTrans server any longer; Seaview can do the same job. How? That is left as an exercise for the reader.

Henni: /* Part 1 — using Clustal Omega in Seaview */

2024-11-11T15:51:40Z

Part 1 — using Clustal Omega in Seaview

← Older revision		Revision as of 17:51, 11 November 2024
Line 14:		Line 14:
	Clustal Omega only comes as a command line interface, but Seaview provides a graphical interface to it. In this exercise (and next week) we will mainly use Seaview to run Clustal Omega. Later in today's exercise, we will also run MUSCLE through Seaview and a couple of other multiple alignment programs via EBI's multiple alignment page: http://www.ebi.ac.uk/Tools/msa/.		Clustal Omega only comes as a command line interface, but Seaview provides a graphical interface to it. In this exercise (and next week) we will mainly use Seaview to run Clustal Omega. Later in today's exercise, we will also run MUSCLE through Seaview and a couple of other multiple alignment programs via EBI's multiple alignment page: http://www.ebi.ac.uk/Tools/msa/.
	<!--		<!--
	ClustalW/X is free to use and available for many computer systems — including Windows and Mac. If you want to install ClustalX on your own machine, it can be downloaded from here: http://www.clustal.org/clustal2/. It can also be used in an online version under EBI's multiple alignment page: ~~http~~://www.ebi.ac.uk/~~Tools~~/msa/.		ClustalW/X is free to use and available for many computer systems — including Windows and Mac. If you want to install ClustalX on your own machine, it can be downloaded from here: http://www.clustal.org/clustal2/. It can also be used in an online version under EBI's multiple alignment page: https://www.ebi.ac.uk/jdispatcher/msa.

	However, as mentioned during the Multiple Alignment lecture, ClustalW/X is neither the fastest nor the most precise program available. No single program performs best in all comparisons, but one of the best and fastest is called MAFFT. It can be downloaded from here: http://mafft.cbrc.jp/alignment/software/ (Note: command line interface only) or used online from here: http://mafft.cbrc.jp/alignment/server/. It is also available in an online version under		However, as mentioned during the Multiple Alignment lecture, ClustalW/X is neither the fastest nor the most precise program available. No single program performs best in all comparisons, but one of the best and fastest is called MAFFT. It can be downloaded from here: http://mafft.cbrc.jp/alignment/software/ (Note: command line interface only) or used online from here: http://mafft.cbrc.jp/alignment/server/. It is also available in an online version under

WikiSysop: /* Step 8 */

2024-03-15T10:10:57Z

Step 8

← Older revision		Revision as of 12:10, 15 March 2024
Line 188:		Line 188:
	The data set for this part of the exercise will be a cleaned-up version of the Insulin FASTA file from Step 4 above (redundancy reduced and with short informative names), available via [https://teaching.healthtech.dtu.dk/material/22111/Insulin_processed.fasta this link].		The data set for this part of the exercise will be a cleaned-up version of the Insulin FASTA file from Step 4 above (redundancy reduced and with short informative names), available via [https://teaching.healthtech.dtu.dk/material/22111/Insulin_processed.fasta this link].

	Link to the RevTrans 2.0 server: https://services.healthtech.dtu.dk/~~service~~.~~php?RevTrans~~		Link to the RevTrans 2.0 server: https://services.healthtech.dtu.dk/services/RevTrans-2.0/

	Notice that it's possible to specify which translation table to use. You may recognize the options from ''VirtualRibosome''. This is no coincidence: both servers are using the same underlying algorithm for translating DNA to protein.		Notice that it's possible to specify which translation table to use. You may recognize the options from ''VirtualRibosome''. This is no coincidence: both servers are using the same underlying algorithm for translating DNA to protein.

WikiSysop: /* Step 8 */

2024-03-15T10:10:39Z

Step 8

← Older revision		Revision as of 12:10, 15 March 2024
Line 186:		Line 186:
	'''If you haven't read [https://www.ncbi.nlm.nih.gov/pmc/articles/PMC169015/ the RevTrans paper] yet — please quickly skim through it now''' (it's an easy read). The paper explains the concept behind the RevTrans method in details (DNA → Protein; Multiple alignment of the proteins; Construct DNA alignment from the DNA sequences using the peptide alignment as a scaffold).		'''If you haven't read [https://www.ncbi.nlm.nih.gov/pmc/articles/PMC169015/ the RevTrans paper] yet — please quickly skim through it now''' (it's an easy read). The paper explains the concept behind the RevTrans method in details (DNA → Protein; Multiple alignment of the proteins; Construct DNA alignment from the DNA sequences using the peptide alignment as a scaffold).

	The data set for this part of the exercise will be a cleaned-up version of the Insulin FASTA file from Step 4 above (redundancy reduced and with short informative names), available via [~~[Media~~:~~insulin_processed~~.fasta\|this link]].		The data set for this part of the exercise will be a cleaned-up version of the Insulin FASTA file from Step 4 above (redundancy reduced and with short informative names), available via [https://teaching.healthtech.dtu.dk/material/22111/Insulin_processed.fasta this link].

	Link to the RevTrans 2.0 server: https://services.healthtech.dtu.dk/service.php?RevTrans		Link to the RevTrans 2.0 server: https://services.healthtech.dtu.dk/service.php?RevTrans

WikiSysop: /* Step 4 */

2024-03-15T10:09:00Z

Step 4

← Older revision		Revision as of 12:09, 15 March 2024
Line 114:		Line 114:

	===Step 4===		===Step 4===
	'''New data set: Insulin'''. [~~[Media~~:~~insulin_raw~~.fasta\|This FASTA file]] contains the DNA sequence from the Insulin gene from a range of organisms		'''New data set: Insulin'''. [https://teaching.healthtech.dtu.dk/material/22111/Insulin_raw.fasta This FASTA file]] contains the DNA sequence from the Insulin gene from a range of organisms

	Notice that this FASTA file has been auto-generated from a database, and it is currently not that informative with regards to entry names. Before we carry on with the analysis, you need to figure out '''which organisms''' the sequences belong to by '''looking up the entries in GenBank'''. Based on this information construct a '''new FASTA file''' with names that 1) describe what organism each sequence came from and 2) keep in the GenBank ID for later reference.		Notice that this FASTA file has been auto-generated from a database, and it is currently not that informative with regards to entry names. Before we carry on with the analysis, you need to figure out '''which organisms''' the sequences belong to by '''looking up the entries in GenBank'''. Based on this information construct a '''new FASTA file''' with names that 1) describe what organism each sequence came from and 2) keep in the GenBank ID for later reference.

WikiSysop: /* Step 3 */

2024-03-15T10:07:41Z

Step 3

← Older revision		Revision as of 12:07, 15 March 2024
Line 104:		Line 104:
	===Step 3===		===Step 3===
	Now translate the DNA sequences to protein sequences and construct a new alignment.		Now translate the DNA sequences to protein sequences and construct a new alignment.
	Link: [https://services.healthtech.dtu.dk/~~service~~.~~php?VirtualRibosome~~ VirtualRibosome]		Link: [https://services.healthtech.dtu.dk/services/VirtualRibosome-2.0/ VirtualRibosome]

	[[Image:Office-notes-line_drawing.png\|30px\|left]]		[[Image:Office-notes-line_drawing.png\|30px\|left]]

WikiSysop: /* Part 1 — using Clustal Omega in Seaview */

2024-03-15T10:06:56Z

Part 1 — using Clustal Omega in Seaview

← Older revision		Revision as of 12:06, 15 March 2024
Line 57:		Line 57:
	Copy each DNA sequence into your text editor as you find it. Give it a descriptive name that conveys which '''organism''' it comes from, and what '''type of alpha-globin''' it is. '''Remember to save your work often'''.		Copy each DNA sequence into your text editor as you find it. Give it a descriptive name that conveys which '''organism''' it comes from, and what '''type of alpha-globin''' it is. '''Remember to save your work often'''.

	'''Important:''' Always use the gene/protein name ''from the CDS feature'', not the GenBank entry name. (There is a trap buried somewhere here, where the entry name is directly misleading). See this PDF: [~~[Media~~:MultiGeneScreenshot-en.pdf\|How to locate CDS names in GenBank]].		'''Important:''' Always use the gene/protein name ''from the CDS feature'', not the GenBank entry name. (There is a trap buried somewhere here, where the entry name is directly misleading). See this PDF: [https://teaching.healthtech.dtu.dk/material/22111/MultiGeneScreenshot-en.pdf How to locate CDS names in GenBank].

	[[Image:Office-notes-line_drawing.png\|30px\|left]]		[[Image:Office-notes-line_drawing.png\|30px\|left]]