-
Notifications
You must be signed in to change notification settings - Fork 8
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
DrugBank Drug Interaction Sentences Seem to Be of a Standard Type #383
Comments
Here are the most common phrases in the drug interaction sentences:
|
The risk/severity items can be decoded with this code. This maps 1,209,578 of the 2,839,486 drug interaction sentences (roughly 43%): def risk_or_severity_matches(sentence):
start = "The risk or severity of "
increased = " can be increased when "
decreased = " can be decreased when "
middle2 = " is combined with "
if sentence.startswith(start):
sentence_cut1 = sentence.replace(start, '')
dir_flag = ""
middle1 = ""
if increased in sentence_cut1:
dir_flag = "increased"
middle1 = increased
elif decreased in sentence_cut1:
dir_flag = "decreased"
middle1 = decreased
else:
return None, None
sentence_div1 = sentence_cut1.split(middle1)
risk_item = sentence_div1[0]
sentence_cut2 = sentence_cut1.replace(risk_item + middle1, "")
sentence_div2 = sentence_cut2.split(middle2)
comb_item1 = sentence_div2[0]
comb_item2 = sentence_div2[1].strip('.')
return risk_item, (risk_item, dir_flag, comb_item1, comb_item2)
return None, None These are the different risk/severity items, which we will need to map:
|
The increase/decrease activity items were extracted with this code. These sentences account for about 310,627 of the total 2,839,486 (about 10%). def may_impact_activities(sentence):
increased = " may increase the "
decreased = " may decrease the "
dir_flag = ""
middle1 = ""
middle2 = " activities of "
if increased in sentence and middle2 in sentence:
dir_flag = "increased"
middle1 = increased
elif decreased in sentence and middle2 in sentence:
dir_flag = "decreased"
middle1 = decreased
else:
return None, None
sentence_div1 = sentence.split(middle1)
item1 = sentence_div1[0]
sentence_cut1 = sentence.replace(item1 + middle1, "")
sentence_div2 = sentence_cut1.split(middle2)
activity = sentence_div2[0]
item2 = sentence_div2[1]
return activity, (item1, dir_flag, activity, item2) These are the different increase/decrease activity of items that will need to be mapped:
|
This map covers all of the sentence types in the DrugBank drug interaction field: (the keys are placeholders for the fields that change with each entry) "risk_or_severity_increase": ["The risk or severity of ", DISEASE_KEY, " can be increased when ", MAIN_DRUG_KEY " is combined with ", INTERACTION_DRUG_KEY]
"risk_or_severity_decrease": ["The risk or severity of ", DISEASE_KEY, " can be decreased when ", MAIN_DRUG_KEY " is combined with ", INTERACTION_DRUG_KEY]
"may_impact_activity_increase": [MAIN_DRUG_KEY, " may increase the ", ACTIVITY_KEY, " activities of ", INTERACTION_DRUG_KEY]
"may_impact_activity_decrease": [MAIN_DRUG_KEY, " may decrease the ", ACTIVITY_KEY, " activities of ", INTERACTION_DRUG_KEY]
"therapeutic_efficacy_increase": ["The therapeutic efficacy of ", INTERACTION_DRUG_KEY, " can be increased when used in combination with ", MAIN_DRUG_KEY]
"therapeutic_efficacy_decrease": ["The therapeutic efficacy of ", INTERACTION_DRUG_KEY, " can be decreased when used in combination with ", MAIN_DRUG_KEY]
"higher_serum_level": [MAIN_DRUG_KEY, " may decrease the excretion rate of ", INTERACTION_DRUG_KEY, " which could result in a higher serum level"]
"metabolism_increase": ["The metabolism of ", MAIN_DRUG_KEY, " can be increased when combined with ", INTERACTION_DRUG_KEY]
"metabolism_decrease": ["The metabolism of ", MAIN_DRUG_KEY, " can be decreased when combined with ", INTERACTION_DRUG_KEY]
"serum_concentration_increase": ["The serum concentration of ", MAIN_DRUG_KEY, " can be increased when it is combined with ", INTERACTION_DRUG_KEY]
"serum_concentration_decrease": ["The serum concentration of ", MAIN_DRUG_KEY, " can be decreased when it is combined with ", INTERACTION_DRUG_KEY]
"excretion_rate_increase": [INTERACTION_DRUG_KEY, " may increase the excretion rate of ", MAIN_DRUG_KEY, " which could result in a lower serum level and potentially a reduction in efficacy"]
"absorption_decrease": [INTERACTION_DRUG_KEY, " can cause a decrease in the absorption of ", MAIN_DRUG_KEY, " resulting in a reduced serum concentration and potentially a decrease in efficacy"]
"excretion_increase": ["The excretion of ", MAIN_DRUG_KEY, " can be increased when combined with ", INTERACTION_DRUG_KEY]
"excretion_decrease": ["The excretion of ", MAIN_DRUG_KEY, " can be decreased when combined with ", INTERACTION_DRUG_KEY]
"active_metabolites_increase": ["The serum concentration of the active metabolites of ", INTERACTION_DRUG_KEY, " can be increased when ", INTERACTION_DRUG_KEY, " is used in combination with ", MAIN_DRUG_KEY]
"active_metabolites_decrease": ["The serum concentration of the active metabolites of ", INTERACTION_DRUG_KEY, " can be decreased when ", INTERACTION_DRUG_KEY, " is used in combination with ", MAIN_DRUG_KEY]
"bioavailibility_decrease": ["The bioavailability of ", MAIN_DRUG_KEY, " can be decreased when combined with ", INTERACTION_DRUG_KEY]
"bioavailibility_increase": ["The bioavailability of ", MAIN_DRUG_KEY, " can be increased when combined with ", INTERACTION_DRUG_KEY]
"diagnostic_agent_effectiveness_decrease": [INTERACTION_DRUG_KEY, " may decrease effectiveness of ", MAIN_DRUG_KEY, " as a diagnostic agent"]
"absorption_increase": [MAIN_DRUG_KEY, " can cause an increase in the absorption of ", INTERACTION_DRUG_KEY, " resulting in an increased serum concentration and potentially a worsening of adverse effects"]
"diagnostic_agent_effectiveness_increase": [INTERACTION_DRUG_KEY, " may increase effectiveness of ", MAIN_DRUG_KEY, " as a diagnostic agent"]
"absorption_of_decreased": ["The absorption of ", INTERACTION_DRUG_KEY, " can be decreased when combined with ", MAIN_DRUG_KEY]
"protein_binding_decrease": ["The protein binding of ", INTERACTION_DRUG_KEY, " can be decreased when combined with ", MAIN_DRUG_KEY]
"hypersensitivity_reaction_increase": ["The risk of a hypersensitivity reaction to ", INTERACTION_DRUG_KEY, " is increased when it is combined with ", MAIN_DRUG_KEY]
"serum_concentration_of_active_metabolites": ["The serum concentration of the active metabolites of ", MAIN_DRUG_KEY, " can be reduced when ", MAIN_DRUG_KEY, " is used in combination with ", INTERACTION_DRUG_KEY, " resulting in a loss in efficacy"]
"serum_concentration_increased_in_combination": ["The serum concentration of ", INTERACTION_DRUG_KEY, ", an active metabolite of ", MAIN_DRUG_KEY, ", can be increased when used in combination with ", MAIN_DRUG_KEY]
"serum_concentration_decreased_in_combination": ["The serum concentration of ", INTERACTION_DRUG_KEY, ", an active metabolite of ", MAIN_DRUG_KEY, ", can be decreased when used in combination with ", MAIN_DRUG_KEY] Here are the counts for each sentence type: {
"absorption_decrease": 22367,
"absorption_increase": 555,
"absorption_of_decreased": 648,
"active_metabolites_decrease": 154,
"active_metabolites_increase": 578,
"bioavailibility_decrease": 1844,
"bioavailibility_increase": 170,
"diagnostic_agent_effectiveness_decrease": 1505,
"diagnostic_agent_effectiveness_increase": 20,
"excretion_decrease": 19972,
"excretion_increase": 570,
"excretion_rate_increase": 60388,
"higher_serum_level": 352172,
"hypersensitivity_reaction_increase": 2101,
"may_impact_activity_decrease": 89722,
"may_impact_activity_increase": 220905,
"metabolism_decrease": 295535,
"metabolism_increase": 109532,
"protein_binding_decrease": 2820,
"risk_or_severity_decrease": 986,
"risk_or_severity_increase": 1208592,
"serum_concentration_decrease": 40104,
"serum_concentration_decreased_in_combination": 44,
"serum_concentration_increase": 129870,
"serum_concentration_increased_in_combination": 436,
"serum_concentration_of_active_metabolites": 206,
"therapeutic_efficacy_decrease": 232445,
"therapeutic_efficacy_increase": 45245
} |
02f1cbb contains the code for identifying and extracting information from the different sentence types. A few disclaimers:
|
This should hopefully be a good stepping stone towards #369 (to include other free text information from DrugBank in KG2). |
We might be able to get more precise information from the DrugBank drug-interactions by mapping the standard sentence forms onto edge types.
The text was updated successfully, but these errors were encountered: