//internal static List<(int id, string name, List<string> groups)> alphabets = new List<(int id, string name, List<string> groups)>()
//{
// (0, /*program.string_debug*/($@"Normal",new List<string>(){
// /*program.string_debug*/($@"A", /*program.string_debug*/($@"R", /*program.string_debug*/($@"N", /*program.string_debug*/($@"D", /*program.string_debug*/($@"C", /*program.string_debug*/($@"Q", /*program.string_debug*/($@"E", /*program.string_debug*/($@"G", /*program.string_debug*/($@"H", /*program.string_debug*/($@"I", /*program.string_debug*/($@"L", /*program.string_debug*/($@"K", /*program.string_debug*/($@"M", /*program.string_debug*/($@"F", /*program.string_debug*/($@"P", /*program.string_debug*/($@"S", /*program.string_debug*/($@"T", /*program.string_debug*/($@"W", /*program.string_debug*/($@"Y", /*program.string_debug*/($@"V"
// }),
// (1, /*program.string_debug*/($@"Physicochemical",new List<string>(){
// /*program.string_debug*/($@"AVFPMILW", /*program.string_debug*/($@"DE", /*program.string_debug*/($@"RK", /*program.string_debug*/($@"STYHCNGQ"
// }),
// (2, /*program.string_debug*/($@"Hydrophobicity",new List<string>(){
// /*program.string_debug*/($@"AGILPV", /*program.string_debug*/($@"FYW", /*program.string_debug*/($@"DENQRHSTK", /*program.string_debug*/($@"CM"
// }),
// (3, /*program.string_debug*/($@"UniProtKb",new List<string>(){
// /*program.string_debug*/($@"LAGVIP", /*program.string_debug*/($@"DE", /*program.string_debug*/($@"ST", /*program.string_debug*/($@"RKH", /*program.string_debug*/($@"FYW", /*program.string_debug*/($@"NQ", /*program.string_debug*/($@"CM"
// }),
// (4, /*program.string_debug*/($@"PdbSum",new List<string>(){
// /*program.string_debug*/($@"HKR", /*program.string_debug*/($@"DE", /*program.string_debug*/($@"STNQ", /*program.string_debug*/($@"AVLIM", /*program.string_debug*/($@"FYW", /*program.string_debug*/($@"PG", /*program.string_debug*/($@"C"
// }),
// (5, /*program.string_debug*/($@"Venn",new List<string>(){
// /*program.string_debug*/($@"ILV", /*program.string_debug*/($@"TS", /*program.string_debug*/($@"AGCS", /*program.string_debug*/($@"VPAGCSTDN", /*program.string_debug*/($@"NQ", /*program.string_debug*/($@"HKR", /*program.string_debug*/($@"DE", /*program.string_debug*/($@"CSTNQDEHKRYW", /*program.string_debug*/($@"ILVAMFGP", /*program.string_debug*/($@"DEHKR", /*program.string_debug*/($@"ILVACTMFYWHK", /*program.string_debug*/($@"FYWH", /*program.string_debug*/($@"MC"
// })
//};
internal static double[] pssm_to_vector1(List <info_blast_pssm_entry> pssm, enum_pssm_value_type pssm_value_type, bool normalise_all = false)
{
const string module_name = nameof(info_blast_pssm);
const string method_name = nameof(pssm_to_vector1);
var x = (pssm == null || pssm.Count == 0) ? new[] { 0d } : pssm.Select(a => a.score).ToArray();
if (x == null || x.Length == 0)
{
x = new[] { 0d }
}
;
if (normalise_all)
{
x = normalise_array(x);
}
//var sum = scores.Sum();
//var average = scores.Average();
if (pssm_value_type == enum_pssm_value_type.distances)
{
x = distances(x, normalise_all);
}
else if (pssm_value_type == enum_pssm_value_type.intervals)
{
x = intervals(x, normalise_all);
}
return(x); //(scores, sum, average);
}
internal static List <(string alphabet, List <(string col_aa, int lag, double[] values)> x)> pssm_to_vector20col_DT(List <info_blast_pssm_entry> pssm, int max_lag, enum_pssm_value_type pssm_value_type, bool normalise_col = false, bool normalise_all = false, CancellationTokenSource cts = null)
{
const string method_name = nameof(pssm_to_vector20col_DT);
// join all cols with same amino acid, calc average
//var cols = pssm.Select(a => a.position_aa).Distinct().ToList();
//var cols = /*program.string_debug*/($@"ARNDCQEGHILKMFPSTWYV".ToCharArray(); //pssm.Select(a => a.position_aa).Distinct().ToList();
using var i_cts = new CancellationTokenSource();
if (cts == null)
{
cts = i_cts;
}
var result = new List <(string alphabet, List <(string col_aa, int lag, double[] values)> x)>();
var tasks = new List <Task <(string name, List <(string col_aa, int lag, double[] values)> vector)> >();
var tasks_start_time = DateTime.Now;
foreach (var l_alphabet in feature_calcs.aa_alphabets)
{
var alphabet = l_alphabet;
for (var l_lag = 1; l_lag <= max_lag; l_lag++)
{
var lag = l_lag;
Task <(string name, List <(string col_aa, int lag, double[] values)> vector)> task = Task.Run(() =>
{
if (cts != null && cts.IsCancellationRequested)
{
return(default);
internal static List <(string alphabet, List <(string col_aa, double[] values)> x)> pssm_to_vector20col(List <info_blast_pssm_entry> pssm, enum_pssm_value_type pssm_value_type, bool normalise_col = false, bool normalise_all = false)
{
const string module_name = nameof(info_blast_pssm);
const string method_name = nameof(pssm_to_vector20col);
// join all cols with same amino acid, calc average
//var cols = pssm.Select(a => a.position_aa).Distinct().ToList();
//var cols = /*program.string_debug*/($@"ARNDCQEGHILKMFPSTWYV".ToCharArray(); //pssm.Select(a => a.position_aa).Distinct().ToList();
var result = new List <(string alphabet, List <(string col_aa, double[] values)> x)>();
foreach (var alphabet in feature_calcs.aa_alphabets)
{
var alphabet_groups = alphabet.groups;
var vector = new List <(string col_aa, double[] values)>();
foreach (var c1 in alphabet_groups)
{
double[] x = (pssm == null || pssm.Count == 0) ? new[] { 0d } : pssm.Where(a => c1.group_amino_acids.Contains(a.position_aa, StringComparison.Ordinal)).Select(a => a.score).ToArray();
if (x == null || x.Length == 0)
{
x = new[] { 0d }
}
;
if (normalise_col)
{
x = normalise_array(x);
}
if (x == null || x.Length == 0)
{
x = new[] { 0d }
}
;
if (pssm_value_type == enum_pssm_value_type.distances)
{
x = distances(x, normalise_col);
}
else if (pssm_value_type == enum_pssm_value_type.intervals)
{
x = intervals(x, normalise_col);
}
if (x == null || x.Length == 0)
{
x = new[] { 0d }
}
;
//var sum = x.Sum();
//var average = x.Average();
vector.Add((c1.group_amino_acids, x));
}
if (normalise_all)
{
var values = vector.SelectMany(a => a.values).Distinct().ToList();
var min = values.Min();
var max = values.Max();
vector.ForEach(a => a.values = normalise_array(a.values, min, max));
}
result.Add((alphabet.name, vector));
}
return(result);
}
