chrisjurich

#include <iostream>
#include <set>
#include <map>

bool
is_num(std::string& string_token, std::set<char>& accepted_characters){
    auto accepted_chars = std::map<char,int>();
    for(const auto& character : string_token) {
        if(accepted_characters.find(character) == accepted_characters.end()) {
            if(character < '0' or character > '9') {

chrisjurich

#include <iostream>
#include <vector>


struct Strings : std::vector<std::string> {

    std::vector<std::string> tokens;
    // unfortunately you have to override all of the constructors
    Strings(std::vector<std::string> input_tokens) : tokens(std::move(input_tokens)) {}

chrisjurich

def is_header_file(file_name):
    """Method that checks if the file is a C++ header file"""
    file_ending = file_name.split('.')[-1].lower()
    for ending in "h|hh|hpp|hxx|h++".split('|'):
        if ending == file_ending:
            return True
    return False

def is_source_file(file_name):
    """Method that checks if the file is a C++ source file"""

chrisjurich

def eterna_recalculated(row,scale_max=2.3):
    """Helper method that recalculates the Eterna score for an entry from a dataframe. It will then put the score back into the row. Please note that there is not a 1:1 correspondence between the actual and recalculated scores"""
    assert len(row["target_structure"]) == len(row["sequence"])
    # sometimes there is a fingerprint sequence at the end of the sturcutre, If that is the case it needs to be removed
    sequence = re.sub("AAAGAAACAACAACAACAAC$","",row["sequence"])
    # data_len is the number of data points that will be reviewed
    data_len = min(
            len(row["target_structure"]),
            len(row["SHAPE_data"]), # can probably get rid of this one
            len(sequence),

chrisjurich

def blended_eterna_score(row,scale_max=10):
    """Method that finds the blended, non-binary eterna score for a structure. Takes the row from a pandas df as input"""
    assert len(row["target_structure"]) == len(row["sequence"])
    # sometimes there is a fingerprint sequence at the end of the sturcutre, If that is the case it needs to be removed
    sequence = re.sub("AAAGAAACAACAACAACAAC$","",row["sequence"])
    # data_len is the number of data points that will be reviewed
    data_len = min(
            len(row["target_structure"]),
            len(row["SHAPE_data"]), # can probably get rid of this one
            len(sequence),

chrisjurich

import urllib.request
from html.parser import HTMLParser
from bs4 import BeautifulSoup
import re
import datetime

###################################################################################
####################### HELPER FUNCTIONS ##########################################
def get_date_from_href(href):
    """Helper method that gets date from a hypertext link. Assumes format of MM/DD/YYYY. Raises error if number of matches != 1"""

chrisjurich

#include<tuple>
#include<vector>
#include<map>
#include<iostream>

///////////////////////////////////////////////////////////////////
///////////////////////////////////////////////////////////////////
///////////////////////////////////////////////////////////////////
// NOTE: you have to pass the flag "-std=c++17" to the compiler. 
///////////////////////////////////////////////////////////////////

chrisjurich

from enum import Enum
from abc import ABC, abstractmethod
from multipledispatch import dispatch

class MotifType(Enum):
    SINGLESTRAND = 0
    HELIX = 1
    HAIRPIN = 2 
    JUNCTION = 3

chrisjurich

from scipy.stats import mannwhitneyu

# Source: https://www.biorxiv.org/content/10.1101/2020.06.29.178343v2.full.pdf

def dsci( sequence, target, dms ):
    # TODO remember, have to replace the dead nt's with N 
    assert len( sequence ) == len( target ) and len( target ) == len( dms )
    # first, gotta do the paired/unpaired  
    paired, unpaired = [], []
    for nt, db, val in zip( sequence, target, dms):

chrisjurich

import math

def stat_test( lower_avg, lower_err, upper_avg, upper_err ):
    # not a real funciton but shoudl return a p value that the 
    # lower_avg < upper_avg
    raise TypeError("not implemented")