PipeWorkReferenceManual

From Wiki

(Difference between revisions)
Jump to: navigation, search
Current revision (13:52, 2 December 2021) (edit) (undo)
(GXSimilarProfile)
 
(47 intermediate revisions not shown.)
Line 6: Line 6:
For example if we select '0 all concerned RealExp' in GXManageExperiment it means that the RealExp are simply not tested to determine which Experiment is to address.
For example if we select '0 all concerned RealExp' in GXManageExperiment it means that the RealExp are simply not tested to determine which Experiment is to address.
 +
=General purpose tools=
 +
==PasteToHtmlTable==
 +
PipeWorks create very often [[HtmlTable]] to display their results. It is a simple table with lines and columns. You can check any columns to redisplay it, and you can run the [[DoOnList]] on it.
 +
* To get an HtmlTable from a text you prviously copy intp the clipboard, run the PipeWork 'PasteToHtmlTable', paste into the text area and press Evaluate.
 +
* Lines have to be separated with newlines, words in a line have to be separated with tabs or space (please don't mix, I'm not shure it works well)
 +
* If you want to give a name to your columns, add a first line with all keys as from example
 +
 +
#KeysToUse:GeneName AccessNumber Length Sequence
=GO PipeWork Reference Manual=
=GO PipeWork Reference Manual=
Line 22: Line 30:
To recall a previously stored result 'Select to recall it' and press Evaluate. I'm not sure it works well if you mix GOs and GENEs etc. So it should be better to enter a non existing Go for example XXXX.
To recall a previously stored result 'Select to recall it' and press Evaluate. I'm not sure it works well if you mix GOs and GENEs etc. So it should be better to enter a non existing Go for example XXXX.
=GX PipeWork Reference Manual=
=GX PipeWork Reference Manual=
 +
Petite remarque concernant les PipeWork GX:
 +
* on drees souvent laliste des probesets mais sans tenir compte de realexp ou experiment ou arraytype concernés. C'est assez diffichile a faire car des fois on a beaucoup de ps et d'autres fois que quelques unes.
 +
Faudrait-il faire un truc du genre on a plein de parametres Ex, Re, At, Ps, ... avec à chaue fois des listes possibles et surtout du vide. Quelle sera le type de requête SQL à créer pour faire au plus vite à l'exécution ? (actuellemtn on boucle sur tout et on élimien en fin de compte ceux qui ne sont pas concernés.
 +
==GxHaploex==
 +
The PipeWork GXInternalRatioHaploexA908Unlimited extracts the SignalIntensities > 40 with all possible ratios (in fact 0<B/A !!!)
 +
(notice that we get also the downregulated genes, at the end of the sorted ratio list)
 +
 +
The PipeWork creates a file containing the .csv with all probesets. (we renamed it A908.csv)
 +
 +
This file is then used by GxFun::Haploex
 +
 +
# GxFun::InterRatio reads this file and sort it using the
 +
 +
==GXSignalIntensity==
 +
* The first function is to get all signal intensities in one/several condition(s) from a probeset/gene (list).
 +
 +
* The second function is to filter probesets/genes according to their signal intensity from a proseset/gene list or from all probesets/genes of one experiment/chip.
 +
 +
We can get, for each selected condition, all genes:
 +
 +
- expressed or not (tag T or F)
 +
 +
- with a signal intensity superior to a selected minimum
 +
 +
- with a signal intensity inferior to a selected maximum
 +
 +
NB: if several conditions have been selected, we get selected genes by filtering for each condition independently of others. So the output list is certainly redundant.
 +
 +
===How to run the PipeWork GXSimilarProfile===
 +
In the pipework let all values unchanged except
 +
* add list of probeset or genename or keep empty
 +
* select the analysis software
 +
* select the realexp
 +
* select the experiment
 +
* optional: Expressed, minimum, maximum
 +
* select the limit
 +
* Press the "Evaluate and Keep Your Entries"
 +
 +
NB: Arraytype and Samples fields are useless.
 +
 +
==GXSimilarProfile==
 +
The PipeWork GXSimilarProfile allows to get all probeset which have the same expresson profile as a given reference probeset.
 +
 +
# we determine the maximum signal (maxRef) for all conditions for the reference probeset
 +
# we determine the maximum signal (maxCur) for all conditions foreach probeset
 +
# we normalise the reference signals with siRef = siRef/maxRef
 +
# we normalise each signal with siCur = siCur/maxCur
 +
# we reject each probeset if $siCur < (1.-$t)*$siRef or (1.+$t)*$siRef < $siCur
 +
 +
===How to run the PipeWork GXSimilarProfile===
 +
In the pipework let all values unchanged except
 +
* select the analysis software
 +
* select one experiment then rebuild or save ... the sample list will be updated
 +
* select at least 2 samples you want to be with similar profiles
 +
* choose a percentage of variation you allow (0.1 is 10%)
 +
* Press the "Evaluate and Keep Your Entries"
 +
 +
To do : J'ai créé des probesets virtuelles (PsForNxnl1 par exemple). On ne les trouve donc pas quand on part d'une vraie probeset.
 +
 +
==GXInternalRatio==
 +
The PipeWork GXInternalRatio allows to extract all SignalIntensities verifying the conditions and where a list of ratios are in a specific range.
 +
 +
The ratios are internal this means that in one given realexp the ratio between two samples are considered.
 +
 +
Imagine you have an experimet with 3 time points T1 T2 T3 for the wt and the mutant.
 +
 +
GXInternalRatio allows you to extract the probesets for which you have for example
 +
 +
Signal(wt,T1)/Sign(mutant,T1) > 2.8 and Signal(wt,T1)/Sign(wt,T3) > 2.3
 +
 +
You can add as many conditions you want as soon you compare the ratio of the signal intensity of the same probeset for different samples.
 +
 +
===How to run the PipeWork GXInternalRatio===
 +
In the pipework let all values unchanged except
 +
* select the analysis software
 +
* select one experiment
 +
* for each GXInternalRatioChooser in the pipework (put the more restrictive condition in first position to speed up the execution)
 +
** Select 2 samples (with the <Ctrl> button) The first one is called A the second B
 +
** Enter an expression as A/B>3.5 using only 'A/B' 'B/A' 'A-B' 'B-A' '<' '>' and numbers as 3.14 (or start it with # if you want to ignore this chooser)
 +
* Rebuild or Save the PipeWork (this is necessary to propagate the experiment to the line notated as "Please rebuild ...")
 +
* Press the "Evaluate and Keep Your Entries"
 +
 +
==GXRatio==
 +
Enter a Genelist or keep empty, select several RatioTypes or RealExpCondition or RealExpReference or several AnalysisSoftwares and press Evaluate.
 +
Then you get the ratio values which verify all these conditions.
 +
 +
* If you want all ratios of a genelist select '0 all concerned Ratiotype', '0 all concerned RealExpCon', etc.
 +
* If you want the genes (probesets) for a given list of RatioTypes (or RealExpCon, etc) select your favorite RatioTypes (or RealExpCon, etc) and keep the genelist empty.
 +
 +
You can also choose the ratio characteritics interval (Minimum, Maximum). If Minimum is greater the Maximum we select the outer parts of the interval.
 +
==GXSamples==
==GXSamples==
This PipeWork shows nicely all information concerning Samples.
This PipeWork shows nicely all information concerning Samples.
You can select Samples directly within the list of available samples or through Experiments, RealExps and/or Arraytype
You can select Samples directly within the list of available samples or through Experiments, RealExps and/or Arraytype
-
==GxManage ... something=
+
 
 +
==GXDiffExpression==
 +
Enter a Genelist or keep empty, select several RatioTypes or RealExpCondition or RealExpReference or several AnalysisSoftwares and press Evaluate.
 +
Then you get the statistically differentially expressed genes which verify all these conditions.
 +
 
 +
* If you want the differentially expressed genes (probesets) for a given list of RatioTypes (or REalExpCon, etc) select your favorite RatioTypes (or REalExpCon, etc) and keep the genelist empty.
 +
* If you want to know if your genes are differentially expressed in a particular condition, give a GeneList and choose the interesting RatioTypes (or REalExpCon, etc).
 +
 
 +
==GxManage ... something==
All available GXManageXXX mainly allow to update or create new records in GxDb (you can also use the '''Show''')
All available GXManageXXX mainly allow to update or create new records in GxDb (you can also use the '''Show''')
Line 56: Line 163:
* GXManageSample allows to show, update or create a new Treatmenttype in GxD
* GXManageSample allows to show, update or create a new Treatmenttype in GxD
To run one of these Managing tool, select, update or create any missing entities, fill in the textual fields and evaluate the whole PipeWork or run separately it by clicking on the corresponding PWI button.
To run one of these Managing tool, select, update or create any missing entities, fill in the textual fields and evaluate the whole PipeWork or run separately it by clicking on the corresponding PWI button.
 +
==ImAnno==
 +
ImAnno (Image Annotation) provides a set of PipeWorks running on lists of genes with or without annotation
 +
===ImAnnoCorrelateTissues===
 +
Nearly the same as ImAnnoCorrelateListsOfGenes but without an additional list of genes
 +
===ImAnnoCorrelateListsOfGenes===
 +
This PipeWork allows the correlation of Lists of genes.
 +
 +
These lists of genes can be obtained as result of Sieves, the WMS of Tissues, KUROV patterns or even a list of genenames.
 +
 +
# Select some Sieves
 +
# Select some Tissues
 +
# Select som KUROV patterns
 +
#Copy-paste a list of genes separated by space, comma, semicolumn or new-line
 +
# Press 'Evaluate and keep your entries'
 +
You'll get a html table where the pourcentage of correlation of sieve A with sieve B is nAB/(nA+nB+nAB) (nA nbr of genes in A, nAB in (A and B))
 +
 +
You ca also get the correlation matrix obtained by a Spearman Correlation
 +
 +
===ImAnnoCluspackDisplay===
===ImAnnoCluspackDisplay===
This PipeWork allows to display the content of Cluspack data.
This PipeWork allows to display the content of Cluspack data.
Line 61: Line 187:
# Select the Cluspack you want to dispaly and
# Select the Cluspack you want to dispaly and
# Press 'Evaluate and keep your entries'
# Press 'Evaluate and keep your entries'
 +
 +
==StringInteractome==
 +
===StringInteractomeDisplay===
 +
The StringIneractomeDisplay PipeWork allows you to
 +
* list the existing StringInteractome
 +
* select one of them and display it
 +
* rename an existing StringInteractome
 +
* delete an existing StringInteractome
 +
 +
Line 0 indicates which directory is concerned. If you change this line you have to REBUILD the PipeWork so that the select list line 2 is up to date.
 +
 +
If you press "Evaluate And Keep Your Entries" you will see all about the StringInteractome ... and you there you'll use CytoscapeWeb

Current revision

PipeWorkReferenceManual describes each PipeWork provided by Genoret Database

Contents

General behaviour

the SelectFrom_listOfSomething or SelectMultipleFrom_listOfSomething asks the user to select one or more lines from ... Something.

In most of all cases the first line is '0 all concerned Something'. This means that in the processes it will concern no or all Something depeding on the context.

For example if we select '0 all concerned RealExp' in GXManageExperiment it means that the RealExp are simply not tested to determine which Experiment is to address.

General purpose tools

PasteToHtmlTable

PipeWorks create very often HtmlTable to display their results. It is a simple table with lines and columns. You can check any columns to redisplay it, and you can run the DoOnList on it.

  • To get an HtmlTable from a text you prviously copy intp the clipboard, run the PipeWork 'PasteToHtmlTable', paste into the text area and press Evaluate.
  • Lines have to be separated with newlines, words in a line have to be separated with tabs or space (please don't mix, I'm not shure it works well)
  • If you want to give a name to your columns, add a first line with all keys as from example
 #KeysToUse:GeneName AccessNumber Length Sequence

GO PipeWork Reference Manual

  • Actually we manage GO, Genes and Pfam
  • You can get one type from an other. If you start from GO you can specify Down (or a positiv n) Up (or a negative n) or Here (0)
  • You can store the result in a Store. For that change the 'GoStore*' to GoStoreMyBeautifulName
  • You can recall these store with the select (it displays all stores starting with GoStore* (if you didn't change it)).

Go

The Go PipeWork

  1. select within the 'From Go Pfam or Gene' (ie gene)
  2. Enter the 'From list' (ie ion metal binding)
  3. Select 'What to display' (ie GENEsymbol,fullname)
  4. Enter the Level (ie Down to include all children, Up to include parents, Here or enter a number -2 Up, 0 Here, (+n skips 0))
  5. Evaluate

You can also store the result, for that enter a name and click on the button from the line 'Click to store the result' To recall a previously stored result 'Select to recall it' and press Evaluate. I'm not sure it works well if you mix GOs and GENEs etc. So it should be better to enter a non existing Go for example XXXX.

GX PipeWork Reference Manual

Petite remarque concernant les PipeWork GX:

  • on drees souvent laliste des probesets mais sans tenir compte de realexp ou experiment ou arraytype concernés. C'est assez diffichile a faire car des fois on a beaucoup de ps et d'autres fois que quelques unes.

Faudrait-il faire un truc du genre on a plein de parametres Ex, Re, At, Ps, ... avec à chaue fois des listes possibles et surtout du vide. Quelle sera le type de requête SQL à créer pour faire au plus vite à l'exécution ? (actuellemtn on boucle sur tout et on élimien en fin de compte ceux qui ne sont pas concernés.

GxHaploex

The PipeWork GXInternalRatioHaploexA908Unlimited extracts the SignalIntensities > 40 with all possible ratios (in fact 0<B/A !!!) (notice that we get also the downregulated genes, at the end of the sorted ratio list)

The PipeWork creates a file containing the .csv with all probesets. (we renamed it A908.csv)

This file is then used by GxFun::Haploex

  1. GxFun::InterRatio reads this file and sort it using the

GXSignalIntensity

  • The first function is to get all signal intensities in one/several condition(s) from a probeset/gene (list).
  • The second function is to filter probesets/genes according to their signal intensity from a proseset/gene list or from all probesets/genes of one experiment/chip.

We can get, for each selected condition, all genes:

- expressed or not (tag T or F)

- with a signal intensity superior to a selected minimum

- with a signal intensity inferior to a selected maximum

NB: if several conditions have been selected, we get selected genes by filtering for each condition independently of others. So the output list is certainly redundant.

How to run the PipeWork GXSimilarProfile

In the pipework let all values unchanged except

  • add list of probeset or genename or keep empty
  • select the analysis software
  • select the realexp
  • select the experiment
  • optional: Expressed, minimum, maximum
  • select the limit
  • Press the "Evaluate and Keep Your Entries"

NB: Arraytype and Samples fields are useless.

GXSimilarProfile

The PipeWork GXSimilarProfile allows to get all probeset which have the same expresson profile as a given reference probeset.

  1. we determine the maximum signal (maxRef) for all conditions for the reference probeset
  2. we determine the maximum signal (maxCur) for all conditions foreach probeset
  3. we normalise the reference signals with siRef = siRef/maxRef
  4. we normalise each signal with siCur = siCur/maxCur
  5. we reject each probeset if $siCur < (1.-$t)*$siRef or (1.+$t)*$siRef < $siCur

How to run the PipeWork GXSimilarProfile

In the pipework let all values unchanged except

  • select the analysis software
  • select one experiment then rebuild or save ... the sample list will be updated
  • select at least 2 samples you want to be with similar profiles
  • choose a percentage of variation you allow (0.1 is 10%)
  • Press the "Evaluate and Keep Your Entries"

To do : J'ai créé des probesets virtuelles (PsForNxnl1 par exemple). On ne les trouve donc pas quand on part d'une vraie probeset.

GXInternalRatio

The PipeWork GXInternalRatio allows to extract all SignalIntensities verifying the conditions and where a list of ratios are in a specific range.

The ratios are internal this means that in one given realexp the ratio between two samples are considered.

Imagine you have an experimet with 3 time points T1 T2 T3 for the wt and the mutant.

GXInternalRatio allows you to extract the probesets for which you have for example

 Signal(wt,T1)/Sign(mutant,T1) > 2.8  and  Signal(wt,T1)/Sign(wt,T3) > 2.3

You can add as many conditions you want as soon you compare the ratio of the signal intensity of the same probeset for different samples.

How to run the PipeWork GXInternalRatio

In the pipework let all values unchanged except

  • select the analysis software
  • select one experiment
  • for each GXInternalRatioChooser in the pipework (put the more restrictive condition in first position to speed up the execution)
    • Select 2 samples (with the <Ctrl> button) The first one is called A the second B
    • Enter an expression as A/B>3.5 using only 'A/B' 'B/A' 'A-B' 'B-A' '<' '>' and numbers as 3.14 (or start it with # if you want to ignore this chooser)
  • Rebuild or Save the PipeWork (this is necessary to propagate the experiment to the line notated as "Please rebuild ...")
  • Press the "Evaluate and Keep Your Entries"

GXRatio

Enter a Genelist or keep empty, select several RatioTypes or RealExpCondition or RealExpReference or several AnalysisSoftwares and press Evaluate. Then you get the ratio values which verify all these conditions.

  • If you want all ratios of a genelist select '0 all concerned Ratiotype', '0 all concerned RealExpCon', etc.
  • If you want the genes (probesets) for a given list of RatioTypes (or RealExpCon, etc) select your favorite RatioTypes (or RealExpCon, etc) and keep the genelist empty.

You can also choose the ratio characteritics interval (Minimum, Maximum). If Minimum is greater the Maximum we select the outer parts of the interval.

GXSamples

This PipeWork shows nicely all information concerning Samples.

You can select Samples directly within the list of available samples or through Experiments, RealExps and/or Arraytype

GXDiffExpression

Enter a Genelist or keep empty, select several RatioTypes or RealExpCondition or RealExpReference or several AnalysisSoftwares and press Evaluate. Then you get the statistically differentially expressed genes which verify all these conditions.

  • If you want the differentially expressed genes (probesets) for a given list of RatioTypes (or REalExpCon, etc) select your favorite RatioTypes (or REalExpCon, etc) and keep the genelist empty.
  • If you want to know if your genes are differentially expressed in a particular condition, give a GeneList and choose the interesting RatioTypes (or REalExpCon, etc).

GxManage ... something

All available GXManageXXX mainly allow to update or create new records in GxDb (you can also use the Show)

For a New, fill in all fields concerning name, description, ... and select the entities to which the foreign keys will point. (some fields are optional but there is no easy way to determine which one). ANd choos '0 all concerned XXX' For a Update leave blank the key you wan't update.

A good way to proceed is to

  1. preselect the Enter, Select which will be mainly constant for what you are doing,
  2. save the GXManageXXX as GXManageXXXTempo
  3. do your New, Update, Clone
  4. delete GXManageXXXTempo

GXManageRealExp

  • GXManageRealExp allows to show, update, clone or create a new RealExp in GxDb
  • You need to fill in all textual fields. Seelect the Arraytype, Sample, Experiment and evaluate

GXManageExperiment

  • GXManageExperiment allows to show, update, clone or create a new Experiment in GxDb
  • You need to fill in all textual fields. You can select several RealExp ... or do it later with GXManageRealExp.

GXManageSamples

A sample is defined by a Tissue, an Individual (defined by an Organism and a Genotype), a Treatment (with Treatmenttype).

All these entities can be shown, updated and created with this PipeWork

  • GXManageSample allows to show, update or create a new Sample in GxDb
  • GXManageSample allows to show, update or create a new Tissue in GxDb
  • GXManageSample allows to show, update or create a new Individual in GxDb
  • GXManageSample allows to show, update or create a new Organism in GxDb
  • GXManageSample allows to show, update or create a new Genotype in GxDb
  • GXManageSample allows to show, update or create a new Treatment in GxDb
  • GXManageSample allows to show, update or create a new Treatmenttype in GxD

To run one of these Managing tool, select, update or create any missing entities, fill in the textual fields and evaluate the whole PipeWork or run separately it by clicking on the corresponding PWI button.

ImAnno

ImAnno (Image Annotation) provides a set of PipeWorks running on lists of genes with or without annotation

ImAnnoCorrelateTissues

Nearly the same as ImAnnoCorrelateListsOfGenes but without an additional list of genes

ImAnnoCorrelateListsOfGenes

This PipeWork allows the correlation of Lists of genes.

These lists of genes can be obtained as result of Sieves, the WMS of Tissues, KUROV patterns or even a list of genenames.

  1. Select some Sieves
  2. Select some Tissues
  3. Select som KUROV patterns
  4. Copy-paste a list of genes separated by space, comma, semicolumn or new-line
  5. Press 'Evaluate and keep your entries'

You'll get a html table where the pourcentage of correlation of sieve A with sieve B is nAB/(nA+nB+nAB) (nA nbr of genes in A, nAB in (A and B))

You ca also get the correlation matrix obtained by a Spearman Correlation


ImAnnoCluspackDisplay

This PipeWork allows to display the content of Cluspack data.

  1. Select your cadre and organ (1 for eye, 2 for teeth, 3 for ear, 4 for eyefundus)
  2. Select the Cluspack you want to dispaly and
  3. Press 'Evaluate and keep your entries'

StringInteractome

StringInteractomeDisplay

The StringIneractomeDisplay PipeWork allows you to

  • list the existing StringInteractome
  • select one of them and display it
  • rename an existing StringInteractome
  • delete an existing StringInteractome

Line 0 indicates which directory is concerned. If you change this line you have to REBUILD the PipeWork so that the select list line 2 is up to date.

If you press "Evaluate And Keep Your Entries" you will see all about the StringInteractome ... and you there you'll use CytoscapeWeb

Personal tools