Connecting Sub-databases - Revision history

Dw34: /* Executive Summary */

2019-05-17T10:42:19Z

Executive Summary

← Older revision		Revision as of 10:42, 17 May 2019
Line 256:		Line 256:
	It is possible that your [[PATHSAMPLE]] database contains many sub-databases not necessarily connected to one another. Therefore, a lot of information is lost when you try to construct disconnectivity graphs.		It is possible that your [[PATHSAMPLE]] database contains many sub-databases not necessarily connected to one another. Therefore, a lot of information is lost when you try to construct disconnectivity graphs.

	To retrieve this information, we should therefore connect these sub-databases. This can be done efficiently by first identifying which minima contained in respective sub-databases are closest to each other, and then trying to connect these first.		To retrieve this information, we should therefore connect these sub-databases. This connection can be done efficiently by first identifying which minima contained in respective sub-databases are closest to each other, and then trying to connect these first.

	To at first identify the sub-databases present in your database, launch the script '''find_connections.sh''' using '''disconnectionDPS''' with the '''PRINTCONNECTED''' keyword included and the argument to '''CONNECTMIN''' set to '''1'''. This produces a list of '''relevant_connected_*''' files, each one representing a sub-database which lists the minima present in it.		To first identify the sub-databases present in your database, launch the script '''find_connections.sh''' using '''disconnectionDPS''' with the '''PRINTCONNECTED''' keyword included and the argument to '''CONNECTMIN''' set to '''1'''. This produces a list of '''relevant_connected_*''' files, each one representing a sub-database which lists the minima present in it.

	Following this step, the script '''connect_sub_databases.sh''' is ~~launched~~. It determines the distances between all of the possible minima in all of the possible sub-databases. It then prioritises calculations in order to connect the maximum number of sub-databases in the minimum possible number of steps, with those of shortest distance being attempted first.		Following this step, the script '''connect_sub_databases.sh''' is used. It determines the distances between all of the possible minima in all of the possible sub-databases. It then prioritises calculations in order to connect the maximum number of sub-databases in the minimum possible number of steps, with those of shortest distance being attempted first.

	~~The outcome to the launch of this latter script is that a~~ number of folders are created, each providing a connected path between two sub-databases. The data from these folders can be merged into the original using the [[PATHSAMPLE]] keyword '''MERGEDB''', thus connecting all of the previously unconnected sub-databases.		A number of folders are created, each providing a connected path between two sub-databases. The data from these folders can be merged into the original using the [[PATHSAMPLE]] keyword '''MERGEDB''', thus connecting all of the previously unconnected sub-databases.

	This methodology is particularly useful for cases where you have a protein with a cofactor and various sites within a pocket that you think the cofactor can attach to. It provides an efficient method to connect these sites within the pocket, having already sampled each.		This methodology is particularly useful for cases where you have a protein with a cofactor and various sites within a pocket that you think the cofactor can attach to. It provides an efficient method to connect these sites within the pocket, having already sampled each.

Dw34 at 10:40, 17 May 2019

2019-05-17T10:40:57Z

Adk44: /* Step 3: Organising Calculations to Attempt */

2019-05-16T13:46:36Z

Step 3: Organising Calculations to Attempt

← Older revision		Revision as of 13:46, 16 May 2019
Line 194:		Line 194:
	The diagrams above list 14 connections which are recommended to be made. What the script therefore does is to create 14 folders for these connections to be attempted. These are named as 00003_02150, 09000_12405 etc.		The diagrams above list 14 connections which are recommended to be made. What the script therefore does is to create 14 folders for these connections to be attempted. These are named as 00003_02150, 09000_12405 etc.

	Within each sub-folder, a [[PATHSAMPLE]] calculation using '''RETAINSP''' is first of all performed. This ensures that only the minima pertaining to the two sub-databases we are trying to connect are included (eg 00003 and 02150 in the case of the sub-folder 00003_02150). Once this is done, the pathdata file is altered so that rather than '''RETAINSP''' we now wish to use the keywords '''CONNECTPAIRS ~~connecfile~~''' and '''CYCLES 1'''.		Within each sub-folder, a [[PATHSAMPLE]] calculation using '''RETAINSP''' is first of all performed. This ensures that only the minima pertaining to the two sub-databases we are trying to connect are included (eg 00003 and 02150 in the case of the sub-folder 00003_02150) and that the numbering scheme in min.data is consistent with when '''CONNECTUNC LOWESTTEST''' was used. Once this is done, the pathdata file is altered so that rather than '''RETAINSP''' we now wish to use the keywords '''CONNECTPAIRS connectfile''' and '''CYCLES 1'''.

	What '''CONNECTPAIRS''' does is to chose specific minima from min.data for connection attempts. This requires an argument which specifies a file to be read in order to list the minima we want to connect. Typically, this argument is '''connectfile''', and therefore we require a file called '''connectfile'''. This file lists all of the minima we wish to connect according to their line number in min.data. As shown above (and copied immediately below) for the connection we wish to make between the sub-databases 00003 and 02150, the minima we are interested in are 2657 and 2663.		What '''CONNECTPAIRS''' does is to chose specific minima from min.data for connection attempts. This requires an argument which specifies a file to be read in order to list the minima we want to connect. Typically, this argument is '''connectfile''', and therefore we require a file called '''connectfile'''. This file lists all of the minima we wish to connect according to their line number in min.data. As shown above (and copied immediately below) for the connection we wish to make between the sub-databases 00003 and 02150, the minima we are interested in are 2657 and 2663.

Adk44: /* Step 3: Organising Calculations to Attempt */

2019-05-16T13:44:20Z

Step 3: Organising Calculations to Attempt

← Older revision		Revision as of 13:44, 16 May 2019
Line 130:		Line 130:
	Using these principles, the sub-databases were therefore connected as follows.		Using these principles, the sub-databases were therefore connected as follows.

	The first line of ~~lowest~~:		The first line of lowest_to_highest_distances_tot:

	00003_02150 2657 2663 connectlowest> Distance: 0.39352080		00003_02150 2657 2663 connectlowest> Distance: 0.39352080

Adk44: /* Step 3: Organising Calculations to Attempt */

2019-05-16T13:43:06Z

Step 3: Organising Calculations to Attempt

← Older revision		Revision as of 13:43, 16 May 2019
Line 91:		Line 91:
	== Step 3: Organising Calculations to Attempt ==		== Step 3: Organising Calculations to Attempt ==

	Clearly, a pair of minima separated by 39.397 is a more feasible calculation to make than one separated by 42.430 or 75.542. The script we have ('''connect_sub_databases''' therefore reorganises distances_tot to list the proposed connections between pairs from shortest distance to longest. This new reorganised file we give the rather unimaginative name of lowest_to_highest_distances_tot.		Clearly, a pair of minima separated by 39.397 is a more feasible calculation to make than one separated by 42.430 or 75.542. The script we have ('''connect_sub_databases''') therefore reorganises distances_tot to list the proposed connections between pairs from shortest distance to longest. This new reorganised file we give the rather unimaginative name of lowest_to_highest_distances_tot.

	The rest of the script is concerned with connecting all of the sub-databases in as efficient a way, using as few steps, as possible. This is probably best illustrated by an example:		The rest of the script is concerned with connecting all of the sub-databases in as efficient a way, using as few steps, as possible. This is probably best illustrated by an example:

Adk44: /* Step 2: RETAINSP and CONNECTUNC LOWESTTEST */

2019-05-16T13:42:42Z