Implements the functions defined in ed4_protein_2nd_structure.hxx. More...

#include "ed4_protein_2nd_structure.hxx"
#include "ed4_class.hxx"
#include "ed4_awars.hxx"
#include <aw_awar.hxx>
#include <aw_msg.hxx>
#include <aw_root.hxx>
#include "arbdbt.h"
#include <iostream>
#include <awt_config_manager.hxx>

Include dependency graph for ED4_protein_2nd_structure.cxx:

Go to the source code of this file.

Classes
struct	pfold_mem_handler

Macros
#define	e4_assert(bed) arb_assert(bed)

Functions
int	ED4_pfold_round_sym (double d)
	Symmetric arithmetic rounding of a double value to an integer value. More...

static void	ED4_pfold_init_statics ()
	Initializes static variables. More...

static void	ED4_pfold_find_nucleation_sites (const unsigned char sequence, char structure, int length, const PFOLD_STRUCTURE s)
	Finds nucleation sites that initiate the specified structure. More...

static void	ED4_pfold_extend_nucleation_sites (const unsigned char sequence, char structure, int length, const PFOLD_STRUCTURE s)
	Extends the found nucleation sites in both directions. More...

static void	ED4_pfold_find_turns (const unsigned char sequence, char structure, int length)
	Predicts beta-turns from the given amino acid sequence. More...

static void	ED4_pfold_resolve_overlaps (const unsigned char sequence, char structures[4], int length)
	Resolves overlaps of predicted secondary structures and creates structure summary. More...

static GB_ERROR	ED4_pfold_predict_structure (const unsigned char sequence, char structures[4], int length)
	Predicts protein secondary structures from the amino acid sequence. More...

GB_ERROR	ED4_pfold_calculate_secstruct_match (const unsigned char structure_sai, const unsigned char structure_cmp, const int start, const int end, char *result_buffer, PFOLD_MATCH_METHOD match_method)
	Compares a protein secondary structure with a primary structure or another secondary structure. More...

GB_ERROR	ED4_pfold_set_SAI (char *protstruct, GBDATA gb_main, const char alignment_name, long protstruct_len)
	Sets the reference protein secondary structure SAI. More...

static void	ED4_pfold_select_SAI_and_update_option_menu (AW_window aww, AW_option_menu_struct oms, bool set_sai)
	Callback function to select the reference protein structure SAI and to update the SAI option menu. More...

static void	setup_pfold_config (AWT_config_definition &cdef)

AW_window *	ED4_pfold_create_props_window (AW_root awr, const WindowCallback refreshCallback)
	Creates the "Protein Match Settings" window. More...

Variables
name_value_pair	pfold_match_type_awars []
	Awars for the match type; binds the PFOLD_MATCH_TYPE to the corresponding awar name. More...

char *	pfold_pair_chars [PFOLD_PAIRS]
	Symbols for the match quality (defined by PFOLD_MATCH_TYPE) as used for match methods SECSTRUCT_SECSTRUCT and SECSTRUCT_SEQUENCE_PREDICT in ED4_pfold_calculate_secstruct_match(). More...

char *	pfold_pairs [PFOLD_PAIRS]
	Match pair definition (see PFOLD_MATCH_TYPE) as used for match methods SECSTRUCT_SECSTRUCT and SECSTRUCT_SEQUENCE_PREDICT in ED4_pfold_calculate_secstruct_match(). More...

static struct pfold_mem_handler	pfold_dealloc

static const char *	amino_acids = "ARDNCEQGHILKMFPSTWYV"
	Specifies the characters used for amino acid one letter code. More...

static int *	char2AA = NULp
	Maps character to amino acid one letter code. More...

static char	structure_chars [3] = { 'H', 'E', 'T' }
	Characters representing protein secondary structure. More...

static const char *	structure_breaker [2]
	Amino acids that break a certain structure (ALPHA_HELIX or BETA_SHEET) as used in ED4_pfold_extend_nucleation_sites(). More...

static const char *	structure_indifferent [2]
	Amino acids that are indifferent for a certain structure (ALPHA_HELIX or BETA_SHEET) as used in ED4_pfold_extend_nucleation_sites(). More...

static name_value_pair	pfold_match_method_awars [4]
	Awars for the match method; binds the PFOLD_MATCH_METHOD to the corresponding name that is used to create the menu in ED4_pfold_create_props_window(). More...

static double	max_former_value [3] = { 1.42, 1.62, 156 }
	Maximum former value for alpha-helix, beta-sheet (in cf_parameters) and beta-turn (in cf_parameters_norm). More...

static double	min_former_value [3] = { 0.0, 0.0, 47 }
	Minimum former value for alpha-helix, beta-sheet (in cf_parameters) and beta-turn (in cf_parameters_norm). More...

static double	max_breaker_value [3] = { 1.21, 2.03, 0.0 }
	Maximum breaker value for alpha-helix, beta-sheet (in cf_parameters) and beta-turn (no breaker values => 0). More...

static double	cf_parameters [20][4]
	Former and breaker values for alpha-helices and beta-sheets (= strands). More...

static double	cf_parameters_norm [20][7]
	Normalized former values for alpha-helices, beta-sheets (= strands) and beta-turns as well as beta-turn probabilities. More...

Detailed Description

Implements the functions defined in ed4_protein_2nd_structure.hxx.

Author: Markus Urban

Date: 2008-02-08

See also: Refer to ed4_protein_2nd_structure.hxx for details, please.

Definition in file ED4_protein_2nd_structure.cxx.

Macro Definition Documentation

#define e4_assert ( bed ) arb_assert(bed)

Definition at line 31 of file ED4_protein_2nd_structure.cxx.

Referenced by ED4_pfold_calculate_secstruct_match(), ED4_pfold_extend_nucleation_sites(), ED4_pfold_find_nucleation_sites(), ED4_pfold_find_turns(), ED4_pfold_predict_structure(), ED4_pfold_resolve_overlaps(), and ED4_pfold_select_SAI_and_update_option_menu().

Function Documentation

int ED4_pfold_round_sym ( double d )

inline

Symmetric arithmetic rounding of a double value to an integer value.

Parameters

[in] d Value to be rounded

Returns: Rounded value

Rounds a double value to an integer value using symmetric arithmetic rounding, i.e. a number $x.y$ is rounded to $x$ if $y < 5$ and to $x+1$ otherwise.

Definition at line 281 of file ED4_protein_2nd_structure.cxx.

Referenced by ED4_pfold_calculate_secstruct_match().

static void ED4_pfold_init_statics ( )

static

Initializes static variables.

So far, this function only concerns char2AA which gets initialized here. See char2AA for details on the values. It is called by ED4_pfold_predict_structure() and ED4_pfold_calculate_secstruct_match().

Attention: If any other prediction function is used alone before calling one of the mentioned functions, this function has to be called first.

Definition at line 295 of file ED4_protein_2nd_structure.cxx.

Referenced by ED4_pfold_calculate_secstruct_match(), and ED4_pfold_predict_structure().

static void ED4_pfold_find_nucleation_sites	(	const unsigned char *	sequence,
		char *	structure,
		int	length,
		const PFOLD_STRUCTURE	s
	)

static

Finds nucleation sites that initiate the specified structure.

Parameters

[in]	sequence	Amino acid sequence
[out]	structure	Predicted secondary structure
[in]	length	Size of sequence and structure
[in]	s	Secondary structure type (either ALPHA_HELIX or BETA_SHEET)

This function finds nucleation sites that initiate the specified structure (alpha-helix or beta-sheet). A window of a fixed size is moved over the sequence and former and breaker values (as defined by cf_parameters) for the amino acids in the window are summed up. If the former values in this region reach a certain value and the breaker values do not exceed a certain limit a nucleation site is formed, i.e. the region is assumed to be the corresponding secondary structure. The result is stored in structure.

Definition at line 324 of file ED4_protein_2nd_structure.cxx.

References ALPHA_HELIX, ED4_root::aw_root, AW_root::awar(), BETA_SHEET, cf_parameters, e4_assert, ED4_AWAR_GAP_CHARS, ED4_ROOT, AW_awar::read_string(), and s.

Referenced by ED4_pfold_predict_structure().

static void ED4_pfold_extend_nucleation_sites	(	const unsigned char *	sequence,
		char *	structure,
		int	length,
		const PFOLD_STRUCTURE	s
	)

static

Extends the found nucleation sites in both directions.

Parameters

[in]	sequence	Amino acid sequence
[out]	structure	Predicted secondary structure
[in]	length	Size of sequence and structure
[in]	s	Secondary structure type (either ALPHA_HELIX or BETA_SHEET)

The function extends the nucleation sites found by ED4_pfold_find_nucleation_sites() in both directions. Extension continues until a certain amino acid constellation is found. The amino acid 'P' breaks an alpha-helix and 'P' as well as 'E' break a beta-sheet. Also, two successive breakers or one breaker followed by an indifferent amino acid (as defined by structure_breaker and structure_indifferent) break the structure. The result is stored in structure.

Definition at line 386 of file ED4_protein_2nd_structure.cxx.

References ALPHA_HELIX, ED4_root::aw_root, AW_root::awar(), BETA_SHEET, e4_assert, ED4_AWAR_GAP_CHARS, ED4_ROOT, length, AW_awar::read_string(), s, and start.

Referenced by ED4_pfold_predict_structure().

static void ED4_pfold_find_turns	(	const unsigned char *	sequence,
		char *	structure,
		int	length
	)

static

Predicts beta-turns from the given amino acid sequence.

Parameters

[in]	sequence	Amino acid sequence
[out]	structure	Predicted secondary structure
[in]	length	Size of sequence and structure

A window of a fixed size is moved over the sequence and former values for alpha-helices, beta-sheets and beta-turns are summed up. In addition, beta-turn probabilities are multiplied. The values are specified in cf_parameters_norm. If the former values for beta-turn are greater than the ones for alpha-helix and beta-sheet and the turn probabilities exceed a certain limit the region is assumed to be a beta-turn. The result is stored in structure.

Definition at line 504 of file ED4_protein_2nd_structure.cxx.

References ED4_root::aw_root, AW_root::awar(), BETA_TURN, cf_parameters_norm, e4_assert, ED4_AWAR_GAP_CHARS, ED4_ROOT, and AW_awar::read_string().

Referenced by ED4_pfold_predict_structure().

static void ED4_pfold_resolve_overlaps	(	const unsigned char *	sequence,
		char *	structures[4],
		int	length
	)

static

Resolves overlaps of predicted secondary structures and creates structure summary.

Parameters

[in]	sequence	Amino acid sequence
[in,out]	structures	Predicted secondary structures (ALPHA_HELIX, BETA_SHEET, BETA_TURN and STRUCTURE_SUMMARY, in this order)
[in]	length	Size of sequence and structures[i]

The function takes the given predicted structures (alpha-helix, beta-sheet and beta-turn) and searches for overlapping regions. If a beta-turn is found the structure summary is assumed to be a beta-turn. For overlapping alpha-helices and beta-sheets the former values are summed up for this region and the structure summary is assumed to be the structure type with the higher former value. The result is stored in structures 3.

Attention: I couldn't find a standard procedure for resolving overlaps and there might be other (possibly better) ways to do that.

Definition at line 573 of file ED4_protein_2nd_structure.cxx.

References ALPHA_HELIX, ED4_root::aw_root, AW_root::awar(), BETA_SHEET, BETA_TURN, cf_parameters, e4_assert, ED4_AWAR_GAP_CHARS, ED4_ROOT, length, AW_awar::read_string(), s, start, and STRUCTURE_SUMMARY.

Referenced by ED4_pfold_predict_structure().

static GB_ERROR ED4_pfold_predict_structure	(	const unsigned char *	sequence,
		char *	structures[4],
		int	length
	)

static

Predicts protein secondary structures from the amino acid sequence.

Parameters

[in]	sequence	Amino acid sequence
[out]	structures	Predicted secondary structures (ALPHA_HELIX, BETA_SHEET, BETA_TURN and STRUCTURE_SUMMARY, in this order)
[in]	length	Size of sequence and structures[i]

Returns: Error description, if an error occurred; 0 otherwise

This function predicts the protein secondary structures from the amino acid sequence according to the Chou-Fasman algorithm. In a first step, nucleation sites for alpha-helices and beta-sheets are found using ED4_pfold_find_nucleation_sites(). In a next step, the found structures are extended obeying certain rules with ED4_pfold_extend_nucleation_sites(). Beta-turns are found with the function ED4_pfold_find_turns(). In a final step, overlapping regions are identified and resolved to create a structure summary with ED4_pfold_resolve_overlaps(). The results are written to structures[i] and can be accessed via the enums ALPHA_HELIX, BETA_SHEET, BETA_TURN and STRUCTURE_SUMMARY.

Definition at line 668 of file ED4_protein_2nd_structure.cxx.

References ALPHA_HELIX, BETA_SHEET, BETA_TURN, e4_assert, ED4_pfold_extend_nucleation_sites(), ED4_pfold_find_nucleation_sites(), ED4_pfold_find_turns(), ED4_pfold_init_statics(), ED4_pfold_resolve_overlaps(), error(), and NULp.

Referenced by ED4_pfold_calculate_secstruct_match().

GB_ERROR ED4_pfold_calculate_secstruct_match	(	const unsigned char *	structure_sai,
		const unsigned char *	structure_cmp,
		int	start,
		int	end,
		char *	result_buffer,
		PFOLD_MATCH_METHOD	match_method = `SECSTRUCT_SEQUENCE`
	)

Compares a protein secondary structure with a primary structure or another secondary structure.

Parameters

[in]	structure_sai	Reference protein structure SAI (secondary structure)
[in]	structure_cmp	Protein structure to compare (primary or secondary structure)
[in]	start	The start of the match computation (visible area in editor)
[in]	end	The end of the match computation (visible area in editor)
[out]	result_buffer	Result buffer for match symbols
[in]	match_method	Method for structure match computation

Returns: Error description, if an error occurred; 0 otherwise

This function compares a protein secondary structure with a primary structure (= amino acid sequence) or another secondary structure depending on match_method.

Match method SECSTRUCT_SECSTRUCT:: Two secondary structures are compared one by one using the criteria defined by pfold_pairs. The match symbols are taken from pfold_pair_chars.

Match method SECSTRUCT_SEQUENCE:: An amino acid sequence is compared with a secondary structure by taking cohesive parts of the structure - gaps in the alignment are skipped - and computing the normalized difference of former and breaker values for this region in the given sequence such that a value from 0 - 100% for the match quality is generated. By dividing this value into steps of 10% it is mapped to the match symbols defined by PFOLD_PAIR_CHARS_2. Note that bends ('S') are assumed to fit everywhere (=> best match symbol), and if a structure is encountered but no corresponding amino acid the worst match symbol is chosen.

Match method SECSTRUCT_SEQUENCE_PREDICT:: An amino acid sequence is compared with a secondary structure using a full prediction of the secondary structure from its sequence via ED4_pfold_predict_structure() and comparing it one by one with the reference structure. Note that not the structure summary is used for comparison, but the individual predicted structure types as returned in structures[4]. The match criteria are defined in pfold_pairs which is searched in ascending order, i.e. good matches first, then the worse ones. If a match is found the corresponding match symbol (as defined by pfold_pair_chars) is chosen. Note that if a structure is encountered but no corresponding amino acid the worst match symbol is chosen.

The match criteria (for SECSTRUCT_SECSTRUCT and SECSTRUCT_SEQUENCE_PREDICT) as well as the match symbols (for all methods) can be adjusted by the user in the "Protein Match Settings" dialog. The result of the match computation (i.e. the match symbols) is written to the result buffer.

Definition at line 690 of file ED4_protein_2nd_structure.cxx.

References ALPHA_HELIX, ARB_strdup(), ED4_root::aw_root, AW_root::awar(), BETA_SHEET, BETA_TURN, cf_breaker, cf_former, e4_assert, ED4_AWAR_GAP_CHARS, ED4_pfold_init_statics(), ED4_pfold_predict_structure(), ED4_pfold_round_sym(), ED4_ROOT, error(), GB_export_error(), length, max_breaker_value, max_former_value, min, min_former_value, name_value_pair::name, NULp, PFOLD_AWAR_PAIR_TEMPLATE, PFOLD_AWAR_SYMBOL_TEMPLATE, PFOLD_AWAR_SYMBOL_TEMPLATE_2, PFOLD_MATCH_METHOD_COUNT, PFOLD_MATCH_TYPE_COUNT, AW_awar::read_string(), SECSTRUCT_SECSTRUCT, SECSTRUCT_SEQUENCE, SECSTRUCT_SEQUENCE_PREDICT, start, STRUCT_NO_MATCH, STRUCT_PERFECT_MATCH, and STRUCT_UNKNOWN.

Referenced by ED4_show_protein_match_on_device().

GB_ERROR ED4_pfold_set_SAI	(	char **	protstruct,
		GBDATA *	gb_main,
		const char *	alignment_name,
		long *	protstruct_len = `NULp`
	)

Sets the reference protein secondary structure SAI.

Parameters

[out]	protstruct	Pointer to reference protein secondary structure SAI
[in]	gb_main	Main database
[in]	alignment_name	Name of the alignment to search for
[out]	protstruct_len	Length of reference protein secondary structure SAI

Returns: Error description, if an error occurred; 0 otherwise

The function searches the database gb_main for the currently selected SAI as defined by PFOLD_AWAR_SELECTED_SAI and assigns the data of the alignment alignment_name to protstruct. If protstruct_len is specified the length of the new reference SAI is stored. The function is used in the editor to initialize the reference protein secondary structure SAI and to update it if the selected SAI is changed in the "Protein Match Settings" dialog. For this purpose it should be called with &ED4_ROOT->protstruct and &ED4_ROOT->protstruct_len.

Definition at line 967 of file ED4_protein_2nd_structure.cxx.

References ED4_root::aw_root, AW_root::awar(), ED4_ROOT, error(), GB_read_string(), GBS_global_string(), GBT_find_SAI(), GBT_find_sequence(), long, NULp, PFOLD_AWAR_ENABLE, PFOLD_AWAR_SELECTED_SAI, AW_awar::read_int(), AW_awar::read_string(), ta, and AW_awar::write_int().

Referenced by ARB_main(), ED4_alignment_length_changed(), ED4_pfold_select_SAI_and_update_option_menu(), and ED4_manager::update_bases_and_rebuild_consensi().

static void ED4_pfold_select_SAI_and_update_option_menu	(	AW_window *	aww,
		AW_option_menu_struct *	oms,
		bool	set_sai
	)

static

Callback function to select the reference protein structure SAI and to update the SAI option menu.

Parameters

[in]	aww	The calling window
[in,out]	oms	The SAI option menu
[in]	set_sai	Specifies if SAI should be updated

The function is called whenever the selected SAI or the SAI filter is changed in the "Protein Match Settings" dialog (see ED4_pfold_create_props_window()). It can be called with set_sai defined to update the reference protein secondary structure SAI in the editor via ED4_pfold_set_SAI() and to update the selection in the SAI option menu. If set_sai is 0 only the option menu is updated. This is necessary if only the SAI filter changed but not the selected SAI.

Definition at line 1011 of file ED4_protein_2nd_structure.cxx.

References aw_message(), ED4_root::aw_root, AW_root::awar(), AW_window::callback(), AW_window::clear_option_menu(), e4_assert, ED4_pfold_set_SAI(), ED4_ROOT, gb_main, GBT_first_SAI(), GBT_get_name_or_description(), GBT_next_SAI(), ED4_root::get_alignment_name(), ED4_root::get_gb_main(), AW_window::insert_default_option(), AW_window::insert_option(), PFOLD_AWAR_SAI_FILTER, PFOLD_AWAR_SELECTED_SAI, ED4_root::protstruct, ED4_root::protstruct_len, AW_awar::read_string(), ta, and AW_window::update_option_menu().

Referenced by ED4_pfold_create_props_window().

static void setup_pfold_config ( AWT_config_definition & cdef )

static

Definition at line 1045 of file ED4_protein_2nd_structure.cxx.

References AWT_config_definition::add(), GBS_global_string(), name_value_pair::name, PFOLD_AWAR_ENABLE, PFOLD_AWAR_MATCH_METHOD, PFOLD_AWAR_PAIR_TEMPLATE, PFOLD_AWAR_SAI_FILTER, PFOLD_AWAR_SELECTED_SAI, PFOLD_AWAR_SYMBOL_TEMPLATE, and PFOLD_AWAR_SYMBOL_TEMPLATE_2.

Referenced by ED4_pfold_create_props_window().

AW_window* ED4_pfold_create_props_window	(	AW_root *	awr,
		const WindowCallback *	refreshCallback
	)

Creates the "Protein Match Settings" window.

Parameters

[in]	awr	Root window
[in]	cb	Callback struct

Returns: Window

The "Protein Match Settings" window allows the user to configure the properties for protein match computation. These settings include turning the match computation on and off (bound to awar PFOLD_AWAR_ENABLE), selecting the reference protein secondary structure SAI (bound to awar PFOLD_AWAR_SELECTED_SAI), choosing the match method (bound to awar PFOLD_AWAR_MATCH_METHOD, see PFOLD_MATCH_METHOD) and the definition of the match pairs (bound to awar PFOLD_AWAR_PAIR_TEMPLATE and pfold_match_type_awars, see PFOLD_MATCH_TYPE and pfold_pairs) as well as the match symbols (bound to awar PFOLD_AWAR_SYMBOL_TEMPLATE and pfold_match_type_awars or PFOLD_AWAR_SYMBOL_TEMPLATE_2, see PFOLD_MATCH_TYPE and pfold_pair_chars or PFOLD_PAIR_CHARS_2). Via a filter (bound to PFOLD_AWAR_SAI_FILTER) the SAIs shown in the option menu can be narrowed down to a selection of SAIs whose names contain the specified string. The callback function ED4_pfold_select_SAI_and_update_option_menu() is bound to the SAI option menu and the SAI filter to update the selected SAI in the editor or the selection in the SAI option menu.

Definition at line 1062 of file ED4_protein_2nd_structure.cxx.

References AW_window::at(), AW_POPDOWN(), ED4_root::aw_root, AW_ROOT_DEFAULT, AW_root::awar(), AWT_insert_config_manager(), ED4_pfold_select_SAI_and_update_option_menu(), ED4_ROOT, makeHelpCallback(), name_value_pair::name, PFOLD_AWAR_ENABLE, PFOLD_AWAR_MATCH_METHOD, PFOLD_AWAR_PAIR_TEMPLATE, PFOLD_AWAR_SAI_FILTER, PFOLD_AWAR_SELECTED_SAI, PFOLD_AWAR_SYMBOL_TEMPLATE, PFOLD_AWAR_SYMBOL_TEMPLATE_2, AW_awar::read_int(), and setup_pfold_config().

Referenced by ED4_popup_helix_or_pfold_window().

Variable Documentation

name_value_pair pfold_match_type_awars[]

Initial value:

= {
    { "Perfect_match", STRUCT_PERFECT_MATCH   },
    { "Good_match",    STRUCT_GOOD_MATCH      },
    { "Medium_match",  STRUCT_MEDIUM_MATCH    },
    { "Bad_match",     STRUCT_BAD_MATCH       },
    { "No_match",      STRUCT_NO_MATCH        },
    { "Unknown_match", STRUCT_UNKNOWN         },
    { NULp,            PFOLD_MATCH_TYPE_COUNT }
}

Awars for the match type; binds the PFOLD_MATCH_TYPE to the corresponding awar name.

Definition at line 39 of file ED4_protein_2nd_structure.cxx.

Referenced by ed4_create_all_awars().

char* pfold_pair_chars[PFOLD_PAIRS]

Initial value:

= {
    ARB_strdup(" "), 
    ARB_strdup("-"), 
    ARB_strdup("~"), 
    ARB_strdup("+"), 
    ARB_strdup("#"), 
    ARB_strdup("?")  
}

Symbols for the match quality (defined by PFOLD_MATCH_TYPE) as used for match methods SECSTRUCT_SECSTRUCT and SECSTRUCT_SEQUENCE_PREDICT in ED4_pfold_calculate_secstruct_match().

Definition at line 50 of file ED4_protein_2nd_structure.cxx.

Referenced by ed4_create_all_awars().

char* pfold_pairs[PFOLD_PAIRS]

Initial value:

= {
    ARB_strdup("HH GG II TT EE BB SS -- -. .."),          
    ARB_strdup("HG HI HS EB ES TS H- G- I- T- E- B- S-"), 
    ARB_strdup("HT GT IT"),                               
    ARB_strdup("ET BT"),                                  
    ARB_strdup("EH BH EG EI"),                            
    ARB_strdup("")                                        
}

Match pair definition (see PFOLD_MATCH_TYPE) as used for match methods SECSTRUCT_SECSTRUCT and SECSTRUCT_SEQUENCE_PREDICT in ED4_pfold_calculate_secstruct_match().

Definition at line 60 of file ED4_protein_2nd_structure.cxx.

Referenced by ed4_create_all_awars().

struct pfold_mem_handler pfold_dealloc

static

const char* amino_acids = "ARDNCEQGHILKMFPSTWYV"

static

Specifies the characters used for amino acid one letter code.

These are the characters that represent amino acids in one letter code. The order is important as the array initializes char2AA which is used to access array elements in the tables cf_parameters and cf_parameters_norm.

Definition at line 86 of file ED4_protein_2nd_structure.cxx.

int* char2AA = NULp

static

Maps character to amino acid one letter code.

This array maps a character to an integer value. It is initialized with the function ED4_pfold_init_statics() which creates an array of the size 256 (for ISO/IEC 8859-1 character encoding). Characters that represent an amino acid get values from 0 to 19 (according to their position in amino_acids) and all others get the value -1. That way, it can be used to get parameters from the tables cf_parameters and cf_parameters_norm or to check if a certain character represents an amino acid.

Definition at line 98 of file ED4_protein_2nd_structure.cxx.

char structure_chars[3] = { 'H', 'E', 'T' }

static

Characters representing protein secondary structure.

Defines the characters representing secondary structure as output by the function ED4_pfold_predict_structure(). According to common standards, these are:
H = alpha-helix,
E = beta-sheet,
T = beta-turn.

Definition at line 109 of file ED4_protein_2nd_structure.cxx.

const char* structure_breaker[2]

static

Initial value:

= {
        "NYPG",
        "PDESGK"
}

Amino acids that break a certain structure (ALPHA_HELIX or BETA_SHEET) as used in ED4_pfold_extend_nucleation_sites().

Definition at line 112 of file ED4_protein_2nd_structure.cxx.

const char* structure_indifferent[2]

static

Initial value:

= {
        "RTSC",
        "RNHA"
}

Amino acids that are indifferent for a certain structure (ALPHA_HELIX or BETA_SHEET) as used in ED4_pfold_extend_nucleation_sites().

Definition at line 118 of file ED4_protein_2nd_structure.cxx.

name_value_pair pfold_match_method_awars[4]

static

Initial value:

= {
    { "Secondary Structure <-> Secondary Structure",        SECSTRUCT_SECSTRUCT        },
    { "Secondary Structure <-> Sequence",                   SECSTRUCT_SEQUENCE         },
    { "Secondary Structure <-> Sequence (Full Prediction)", SECSTRUCT_SEQUENCE_PREDICT },
    { NULp,                                                 PFOLD_MATCH_METHOD_COUNT }
}

Awars for the match method; binds the PFOLD_MATCH_METHOD to the corresponding name that is used to create the menu in ED4_pfold_create_props_window().

Definition at line 124 of file ED4_protein_2nd_structure.cxx.

double max_former_value[3] = { 1.42, 1.62, 156 }

static

Maximum former value for alpha-helix, beta-sheet (in cf_parameters) and beta-turn (in cf_parameters_norm).

Definition at line 131 of file ED4_protein_2nd_structure.cxx.

Referenced by ED4_pfold_calculate_secstruct_match().

double min_former_value[3] = { 0.0, 0.0, 47 }

static

Minimum former value for alpha-helix, beta-sheet (in cf_parameters) and beta-turn (in cf_parameters_norm).

Definition at line 132 of file ED4_protein_2nd_structure.cxx.

Referenced by ED4_pfold_calculate_secstruct_match().

double max_breaker_value[3] = { 1.21, 2.03, 0.0 }

static

Maximum breaker value for alpha-helix, beta-sheet (in cf_parameters) and beta-turn (no breaker values => 0).

Definition at line 133 of file ED4_protein_2nd_structure.cxx.

Referenced by ED4_pfold_calculate_secstruct_match().

double cf_parameters[20][4]

static

Former and breaker values for alpha-helices and beta-sheets (= strands).

Initial value:

{
//   Helix Former   Strand Former   Helix Breaker   Strand Breaker   Amino
//   Value          Value           Value           Value            Acid
// -----------------------------------------------------------------------
   { 1.34,          0.00,           0.00,           0.00 },          // A
   { 0.00,          0.00,           0.00,           0.00 },          // R
   { 0.50,          0.00,           0.00,           1.39 },          // D
   { 0.00,          0.00,           1.03,           0.00 },          // N
   { 0.00,          1.13,           0.00,           0.00 },          // C
   { 1.42,          0.00,           0.00,           2.03 },          // E
   { 1.05,          1.05,           0.00,           0.00 },          // Q
   { 0.00,          0.00,           1.21,           1.00 },          // G
   { 0.50,          0.00,           0.00,           0.00 },          // H
   { 1.02,          1.52,           0.00,           0.00 },          // I
   { 1.14,          1.24,           0.00,           0.00 },          // L
   { 1.09,          0.00,           0.00,           1.01 },          // K
   { 1.37,          1.00,           0.00,           0.00 },          // M
   { 1.07,          1.31,           0.00,           0.00 },          // F
   { 0.00,          0.00,           1.21,           1.36 },          // P
   { 0.00,          0.00,           0.00,           1.00 },          // S
   { 0.00,          1.13,           0.00,           0.00 },          // T
   { 1.02,          1.30,           0.00,           0.00 },          // W
   { 0.00,          1.40,           1.00,           0.00 },          // Y
   { 1.00,          1.62,           0.00,           0.00 }};         // V

The former and breaker values are used to find alpha-helix and beta-sheet nucleation sites in ED4_pfold_find_nucleation_sites() and to resolve overlaps in ED4_pfold_resolve_overlaps(). Addressing the array with the enums ALPHA_HELIX or BETA_SHEET as second index gives the former values and addressing it with ALPHA_HELIX+2 or BETA_SHEET+2 gives the breaker values. The first index is for the amino acid. Use char2AA to convert an amino acid character to the corresponding index.

See also: Refer to the definition in the source code for commented table.

Definition at line 181 of file ED4_protein_2nd_structure.cxx.

Referenced by ED4_pfold_find_nucleation_sites(), and ED4_pfold_resolve_overlaps().

double cf_parameters_norm[20][7]

static

Normalized former values for alpha-helices, beta-sheets (= strands) and beta-turns as well as beta-turn probabilities.

Initial value:

{
//   P(a)  P(b)  P(turn)  f(i)    f(i+1)  f(i+2)  f(i+3)     Amino Acid
// --------------------------------------------------------------------
   { 142,   83,   66,     0.060,  0.076,  0.035,  0.058 },   // A
   {  98,   93,   95,     0.070,  0.106,  0.099,  0.085 },   // R
   { 101,   54,  146,     0.147,  0.110,  0.179,  0.081 },   // D
   {  67,   89,  156,     0.161,  0.083,  0.191,  0.091 },   // N
   {  70,  119,  119,     0.149,  0.050,  0.117,  0.128 },   // C
   { 151,   37,   74,     0.056,  0.060,  0.077,  0.064 },   // E
   { 111,  110,   98,     0.074,  0.098,  0.037,  0.098 },   // Q
   {  57,   75,  156,     0.102,  0.085,  0.190,  0.152 },   // G
   { 100,   87,   95,     0.140,  0.047,  0.093,  0.054 },   // H
   { 108,  160,   47,     0.043,  0.034,  0.013,  0.056 },   // I
   { 121,  130,   59,     0.061,  0.025,  0.036,  0.070 },   // L
   { 116,   74,  101,     0.055,  0.115,  0.072,  0.095 },   // K
   { 145,  105,   60,     0.068,  0.082,  0.014,  0.055 },   // M
   { 113,  138,   60,     0.059,  0.041,  0.065,  0.065 },   // F
   {  57,   55,  152,     0.102,  0.301,  0.034,  0.068 },   // P
   {  77,   75,  143,     0.120,  0.139,  0.125,  0.106 },   // S
   {  83,  119,   96,     0.086,  0.108,  0.065,  0.079 },   // T
   { 108,  137,   96,     0.077,  0.013,  0.064,  0.167 },   // W
   {  69,  147,  114,     0.082,  0.065,  0.114,  0.125 },   // Y
   { 106,  170,   50,     0.062,  0.048,  0.028,  0.053 }};  // V

The normalized former values are used to find beta-turns in an amino acid sequence in ED4_pfold_find_turns(). Addressing the array with the enums ALPHA_HELIX, BETA_SHEET or BETA_TURN as second index gives the former values and addressing it with BETA_TURN+i $(1 <= i <= 4)$ gives the turn probabilities. The first index is for the amino acid. Use char2AA to convert an amino acid character to the corresponding index.

See also: Refer to the definition in the source code for commented table.

Definition at line 246 of file ED4_protein_2nd_structure.cxx.

Referenced by ED4_pfold_find_turns().

Classes

Macros

Functions

Variables

Detailed Description

Macro Definition Documentation

Function Documentation

Variable Documentation