diff --git a/.env.prod.example b/.env.prod.example index 8d3ab7aa..217b9c91 100644 --- a/.env.prod.example +++ b/.env.prod.example @@ -16,3 +16,5 @@ POSTGRES_PORT= # MAIL EMAIL_HOST_USER= EMAIL_HOST_PASSWORD= +# MATOMO +MATOMO_SITE_ID diff --git a/envipath/settings.py b/envipath/settings.py index 5a18368b..6fdac345 100644 --- a/envipath/settings.py +++ b/envipath/settings.py @@ -357,3 +357,6 @@ if MS_ENTRA_ENABLED: MS_ENTRA_AUTHORITY = f"https://login.microsoftonline.com/{MS_ENTRA_TENANT_ID}" MS_ENTRA_REDIRECT_URI = os.environ["MS_REDIRECT_URI"] MS_ENTRA_SCOPES = os.environ.get("MS_SCOPES", "").split(",") + +# Site ID 10 -> beta.envipath.org +MATOMO_SITE_ID = os.environ.get("MATOMO_SITE_ID", "10") diff --git a/epdb/admin.py b/epdb/admin.py index fefcdc32..88f851af 100644 --- a/epdb/admin.py +++ b/epdb/admin.py @@ -7,6 +7,7 @@ from .models import ( GroupPackagePermission, Package, MLRelativeReasoning, + EnviFormer, Compound, CompoundStructure, SimpleAmbitRule, @@ -19,11 +20,12 @@ from .models import ( Setting, ExternalDatabase, ExternalIdentifier, + JobLog, ) class UserAdmin(admin.ModelAdmin): - pass + list_display = ["username", "email", "is_active"] class UserPackagePermissionAdmin(admin.ModelAdmin): @@ -38,8 +40,14 @@ class GroupPackagePermissionAdmin(admin.ModelAdmin): pass +class JobLogAdmin(admin.ModelAdmin): + pass + + class EPAdmin(admin.ModelAdmin): search_fields = ["name", "description"] + list_display = ["name", "url", "created"] + ordering = ["-created"] class PackageAdmin(EPAdmin): @@ -50,6 +58,10 @@ class MLRelativeReasoningAdmin(EPAdmin): pass +class EnviFormerAdmin(EPAdmin): + pass + + class CompoundAdmin(EPAdmin): pass @@ -102,8 +114,10 @@ admin.site.register(User, UserAdmin) admin.site.register(UserPackagePermission, UserPackagePermissionAdmin) admin.site.register(Group, GroupAdmin) admin.site.register(GroupPackagePermission, GroupPackagePermissionAdmin) +admin.site.register(JobLog, JobLogAdmin) admin.site.register(Package, PackageAdmin) admin.site.register(MLRelativeReasoning, MLRelativeReasoningAdmin) +admin.site.register(EnviFormer, EnviFormerAdmin) admin.site.register(Compound, CompoundAdmin) admin.site.register(CompoundStructure, CompoundStructureAdmin) admin.site.register(SimpleAmbitRule, SimpleAmbitRuleAdmin) diff --git a/epdb/management/commands/create_ml_models.py b/epdb/management/commands/create_ml_models.py index 8cf3fd55..89fbc0ec 100644 --- a/epdb/management/commands/create_ml_models.py +++ b/epdb/management/commands/create_ml_models.py @@ -7,10 +7,11 @@ from epdb.models import MLRelativeReasoning, EnviFormer, Package class Command(BaseCommand): """This command can be run with - `python manage.py create_ml_models [model_names] -d [data_packages] OPTIONAL: -e [eval_packages]` - For example, to train both EnviFormer and MLRelativeReasoning on BBD and SOIL and evaluate them on SLUDGE - the below command would be used: - `python manage.py create_ml_models enviformer mlrr -d bbd soil -e sludge + `python manage.py create_ml_models [model_names] -d [data_packages] FOR MLRR ONLY: -r [rule_packages] + OPTIONAL: -e [eval_packages] -t threshold` + For example, to train both EnviFormer and MLRelativeReasoning on BBD and SOIL and evaluate them on SLUDGE with a + threshold of 0.6, the below command would be used: + `python manage.py create_ml_models enviformer mlrr -d bbd soil -e sludge -t 0.6 """ def add_arguments(self, parser): @@ -34,6 +35,13 @@ class Command(BaseCommand): help="Rule Packages mandatory for MLRR", default=[], ) + parser.add_argument( + "-t", + "--threshold", + type=float, + help="Model prediction threshold", + default=0.5, + ) @transaction.atomic def handle(self, *args, **options): @@ -67,7 +75,11 @@ class Command(BaseCommand): return packages # Iteratively create models in options["model_names"] - print(f"Creating models: {options['model_names']}") + print(f"Creating models: {options['model_names']}\n" + f"Data packages: {options['data_packages']}\n" + f"Rule Packages (only for MLRR): {options['rule_packages']}\n" + f"Eval Packages: {options['eval_packages']}\n" + f"Threshold: {options['threshold']:.2f}") data_packages = decode_packages(options["data_packages"]) eval_packages = decode_packages(options["eval_packages"]) rule_packages = decode_packages(options["rule_packages"]) @@ -78,9 +90,10 @@ class Command(BaseCommand): pack, data_packages=data_packages, eval_packages=eval_packages, - threshold=0.5, - name="EnviFormer - T0.5", - description="EnviFormer transformer", + threshold=options['threshold'], + name=f"EnviFormer - {', '.join(options['data_packages'])} - T{options['threshold']:.2f}", + description=f"EnviFormer transformer trained on {options['data_packages']} " + f"evaluated on {options['eval_packages']}.", ) elif model_name == "mlrr": model = MLRelativeReasoning.create( @@ -88,9 +101,10 @@ class Command(BaseCommand): rule_packages=rule_packages, data_packages=data_packages, eval_packages=eval_packages, - threshold=0.5, - name="ECC - BBD - T0.5", - description="ML Relative Reasoning", + threshold=options['threshold'], + name=f"ECC - {', '.join(options['data_packages'])} - T{options['threshold']:.2f}", + description=f"ML Relative Reasoning trained on {options['data_packages']} with rules from " + f"{options['rule_packages']} and evaluated on {options['eval_packages']}.", ) else: raise ValueError(f"Cannot create model of type {model_name}, unknown model type") @@ -100,6 +114,6 @@ class Command(BaseCommand): print(f"Training {model_name}") model.build_model() print(f"Evaluating {model_name}") - model.evaluate_model() + model.evaluate_model(False, eval_packages=eval_packages) print(f"Saving {model_name}") model.save() diff --git a/epdb/management/commands/update_job_logs.py b/epdb/management/commands/update_job_logs.py new file mode 100644 index 00000000..a5b17cfa --- /dev/null +++ b/epdb/management/commands/update_job_logs.py @@ -0,0 +1,38 @@ +from datetime import date, timedelta + +from django.core.management.base import BaseCommand +from django.db import transaction + +from epdb.models import JobLog + + +class Command(BaseCommand): + def add_arguments(self, parser): + parser.add_argument( + "--cleanup", + type=int, + default=None, + help="Remove all logs older than this number of days. Default is None, which does not remove any logs.", + ) + + @transaction.atomic + def handle(self, *args, **options): + if options["cleanup"] is not None: + cleanup_dt = date.today() - timedelta(days=options["cleanup"]) + print(JobLog.objects.filter(created__lt=cleanup_dt).delete()) + + logs = JobLog.objects.filter(status="INITIAL") + print(f"Found {logs.count()} logs to update") + updated = 0 + for log in logs: + res = log.check_for_update() + if res: + updated += 1 + + print(f"Updated {updated} logs") + + from django.db.models import Count + + qs = JobLog.objects.values("status").annotate(total=Count("status")) + for r in qs: + print(r["status"], r["total"]) diff --git a/epdb/migrations/0009_joblog.py b/epdb/migrations/0009_joblog.py new file mode 100644 index 00000000..5c731eb1 --- /dev/null +++ b/epdb/migrations/0009_joblog.py @@ -0,0 +1,66 @@ +# Generated by Django 5.2.7 on 2025-10-27 09:39 + +import django.db.models.deletion +import django.utils.timezone +import model_utils.fields +from django.conf import settings +from django.db import migrations, models + + +class Migration(migrations.Migration): + dependencies = [ + ("epdb", "0008_enzymelink"), + ] + + operations = [ + migrations.CreateModel( + name="JobLog", + fields=[ + ( + "id", + models.BigAutoField( + auto_created=True, primary_key=True, serialize=False, verbose_name="ID" + ), + ), + ( + "created", + model_utils.fields.AutoCreatedField( + default=django.utils.timezone.now, editable=False, verbose_name="created" + ), + ), + ( + "modified", + model_utils.fields.AutoLastModifiedField( + default=django.utils.timezone.now, editable=False, verbose_name="modified" + ), + ), + ("task_id", models.UUIDField(unique=True)), + ("job_name", models.TextField()), + ( + "status", + models.CharField( + choices=[ + ("INITIAL", "Initial"), + ("SUCCESS", "Success"), + ("FAILURE", "Failure"), + ("REVOKED", "Revoked"), + ("IGNORED", "Ignored"), + ], + default="INITIAL", + max_length=20, + ), + ), + ("done_at", models.DateTimeField(blank=True, default=None, null=True)), + ("task_result", models.TextField(blank=True, default=None, null=True)), + ( + "user", + models.ForeignKey( + on_delete=django.db.models.deletion.CASCADE, to=settings.AUTH_USER_MODEL + ), + ), + ], + options={ + "abstract": False, + }, + ), + ] diff --git a/epdb/models.py b/epdb/models.py index b7f9e82a..3db8ce0f 100644 --- a/epdb/models.py +++ b/epdb/models.py @@ -2226,10 +2226,18 @@ class PackageBasedModel(EPModel): self.model_status = self.BUILT_NOT_EVALUATED self.save() - def evaluate_model(self, **kwargs): + def evaluate_model(self, multigen: bool, eval_packages: List["Package"] = None, **kwargs): if self.model_status != self.BUILT_NOT_EVALUATED: raise ValueError(f"Can't evaluate a model in state {self.model_status}!") + if multigen: + self.multigen_eval = multigen + self.save() + + if eval_packages is not None: + for p in eval_packages: + self.eval_packages.add(p) + self.model_status = self.EVALUATING self.save() @@ -2526,7 +2534,6 @@ class RuleBasedRelativeReasoning(PackageBasedModel): package: "Package", rule_packages: List["Package"], data_packages: List["Package"], - eval_packages: List["Package"], threshold: float = 0.5, min_count: int = 10, max_count: int = 0, @@ -2575,10 +2582,6 @@ class RuleBasedRelativeReasoning(PackageBasedModel): for p in rule_packages: rbrr.data_packages.add(p) - if eval_packages: - for p in eval_packages: - rbrr.eval_packages.add(p) - rbrr.save() return rbrr @@ -2633,7 +2636,6 @@ class MLRelativeReasoning(PackageBasedModel): package: "Package", rule_packages: List["Package"], data_packages: List["Package"], - eval_packages: List["Package"], threshold: float = 0.5, name: "str" = None, description: str = None, @@ -2673,10 +2675,6 @@ class MLRelativeReasoning(PackageBasedModel): for p in rule_packages: mlrr.data_packages.add(p) - if eval_packages: - for p in eval_packages: - mlrr.eval_packages.add(p) - if build_app_domain: ad = ApplicabilityDomain.create( mlrr, @@ -2953,7 +2951,6 @@ class EnviFormer(PackageBasedModel): def create( package: "Package", data_packages: List["Package"], - eval_packages: List["Package"], threshold: float = 0.5, name: "str" = None, description: str = None, @@ -2986,10 +2983,6 @@ class EnviFormer(PackageBasedModel): for p in data_packages: mod.data_packages.add(p) - if eval_packages: - for p in eval_packages: - mod.eval_packages.add(p) - # if build_app_domain: # ad = ApplicabilityDomain.create(mod, app_domain_num_neighbours, app_domain_reliability_threshold, # app_domain_local_compatibility_threshold) @@ -3082,10 +3075,18 @@ class EnviFormer(PackageBasedModel): args = {"clz": "EnviFormer"} return args - def evaluate_model(self, **kwargs): + def evaluate_model(self, multigen: bool, eval_packages: List["Package"] = None, **kwargs): if self.model_status != self.BUILT_NOT_EVALUATED: raise ValueError(f"Can't evaluate a model in state {self.model_status}!") + if multigen: + self.multigen_eval = multigen + self.save() + + if eval_packages is not None: + for p in eval_packages: + self.eval_packages.add(p) + self.model_status = self.EVALUATING self.save() @@ -3226,7 +3227,7 @@ class EnviFormer(PackageBasedModel): ds = self.load_dataset() n_splits = kwargs.get("n_splits", 20) - shuff = ShuffleSplit(n_splits=n_splits, test_size=0.25, random_state=42) + shuff = ShuffleSplit(n_splits=n_splits, test_size=0.1, random_state=42) # Single gen eval is done in one loop of train then evaluate rather than storing all n_splits trained models # this helps reduce the memory footprint. @@ -3294,7 +3295,7 @@ class EnviFormer(PackageBasedModel): # Compute splits of the collected pathway and evaluate. Like single gen we train and evaluate in each # iteration instead of storing all trained models. for split_id, (train, test) in enumerate( - ShuffleSplit(n_splits=n_splits, test_size=0.25, random_state=42).split(pathways) + ShuffleSplit(n_splits=n_splits, test_size=0.1, random_state=42).split(pathways) ): train_pathways = [pathways[i] for i in train] test_pathways = [pathways[i] for i in test] @@ -3577,3 +3578,53 @@ class Setting(EnviPathModel): self.public = True self.global_default = True self.save() + + +class JobLogStatus(models.TextChoices): + INITIAL = "INITIAL", "Initial" + SUCCESS = "SUCCESS", "Success" + FAILURE = "FAILURE", "Failure" + REVOKED = "REVOKED", "Revoked" + IGNORED = "IGNORED", "Ignored" + + +class JobLog(TimeStampedModel): + user = models.ForeignKey("epdb.User", models.CASCADE) + task_id = models.UUIDField(unique=True) + job_name = models.TextField(null=False, blank=False) + status = models.CharField( + max_length=20, + choices=JobLogStatus.choices, + default=JobLogStatus.INITIAL, + ) + + done_at = models.DateTimeField(null=True, blank=True, default=None) + task_result = models.TextField(null=True, blank=True, default=None) + + def check_for_update(self): + async_res = self.get_result() + new_status = async_res.state + + TERMINAL_STATES = [ + "SUCCESS", + "FAILURE", + "REVOKED", + "IGNORED", + ] + + if new_status != self.status and new_status in TERMINAL_STATES: + self.status = new_status + self.done_at = async_res.date_done + + if new_status == "SUCCESS": + self.task_result = async_res.result + + self.save() + + return True + return False + + def get_result(self): + from celery.result import AsyncResult + + return AsyncResult(str(self.task_id)) diff --git a/epdb/tasks.py b/epdb/tasks.py index b9845c86..b872d4a9 100644 --- a/epdb/tasks.py +++ b/epdb/tasks.py @@ -1,10 +1,15 @@ +import csv +import io import logging -from typing import Optional -from celery.utils.functional import LRUCache -from celery import shared_task -from epdb.models import Pathway, Node, EPModel, Setting -from epdb.logic import SPathway +from datetime import datetime +from typing import Any, Callable, List, Optional +from uuid import uuid4 +from celery import shared_task +from celery.utils.functional import LRUCache + +from epdb.logic import SPathway +from epdb.models import EPModel, JobLog, Node, Package, Pathway, Rule, Setting, User, Edge logger = logging.getLogger(__name__) ML_CACHE = LRUCache(3) # Cache the three most recent ML models to reduce load times. @@ -16,6 +21,40 @@ def get_ml_model(model_pk: int): return ML_CACHE[model_pk] +def dispatch_eager(user: "User", job: Callable, *args, **kwargs): + try: + x = job(*args, **kwargs) + log = JobLog() + log.user = user + log.task_id = uuid4() + log.job_name = job.__name__ + log.status = "SUCCESS" + log.done_at = datetime.now() + log.task_result = str(x) if x else None + log.save() + + return x + except Exception as e: + logger.exception(e) + raise e + + +def dispatch(user: "User", job: Callable, *args, **kwargs): + try: + x = job.delay(*args, **kwargs) + log = JobLog() + log.user = user + log.task_id = x.task_id + log.job_name = job.__name__ + log.status = "INITIAL" + log.save() + + return x.result + except Exception as e: + logger.exception(e) + raise e + + @shared_task(queue="background") def mul(a, b): return a * b @@ -33,17 +72,55 @@ def send_registration_mail(user_pk: int): pass -@shared_task(queue="model") -def build_model(model_pk: int): +@shared_task(bind=True, queue="model") +def build_model(self, model_pk: int): mod = EPModel.objects.get(id=model_pk) - mod.build_dataset() - mod.build_model() + + if JobLog.objects.filter(task_id=self.request.id).exists(): + JobLog.objects.filter(task_id=self.request.id).update(status="RUNNING", task_result=mod.url) + + try: + mod.build_dataset() + mod.build_model() + except Exception as e: + if JobLog.objects.filter(task_id=self.request.id).exists(): + JobLog.objects.filter(task_id=self.request.id).update( + status="FAILED", task_result=mod.url + ) + + raise e + + if JobLog.objects.filter(task_id=self.request.id).exists(): + JobLog.objects.filter(task_id=self.request.id).update(status="SUCCESS", task_result=mod.url) + + return mod.url -@shared_task(queue="model") -def evaluate_model(model_pk: int): +@shared_task(bind=True, queue="model") +def evaluate_model(self, model_pk: int, multigen: bool, package_pks: Optional[list] = None): + packages = None + + if package_pks: + packages = Package.objects.filter(pk__in=package_pks) + mod = EPModel.objects.get(id=model_pk) - mod.evaluate_model() + if JobLog.objects.filter(task_id=self.request.id).exists(): + JobLog.objects.filter(task_id=self.request.id).update(status="RUNNING", task_result=mod.url) + + try: + mod.evaluate_model(multigen, eval_packages=packages) + except Exception as e: + if JobLog.objects.filter(task_id=self.request.id).exists(): + JobLog.objects.filter(task_id=self.request.id).update( + status="FAILED", task_result=mod.url + ) + + raise e + + if JobLog.objects.filter(task_id=self.request.id).exists(): + JobLog.objects.filter(task_id=self.request.id).update(status="SUCCESS", task_result=mod.url) + + return mod.url @shared_task(queue="model") @@ -52,9 +129,13 @@ def retrain(model_pk: int): mod.retrain() -@shared_task(queue="predict") +@shared_task(bind=True, queue="predict") def predict( - pw_pk: int, pred_setting_pk: int, limit: Optional[int] = None, node_pk: Optional[int] = None + self, + pw_pk: int, + pred_setting_pk: int, + limit: Optional[int] = None, + node_pk: Optional[int] = None, ) -> Pathway: pw = Pathway.objects.get(id=pw_pk) setting = Setting.objects.get(id=pred_setting_pk) @@ -65,6 +146,9 @@ def predict( pw.kv.update(**{"status": "running"}) pw.save() + if JobLog.objects.filter(task_id=self.request.id).exists(): + JobLog.objects.filter(task_id=self.request.id).update(status="RUNNING", task_result=pw.url) + try: # regular prediction if limit is not None: @@ -89,7 +173,111 @@ def predict( except Exception as e: pw.kv.update({"status": "failed"}) pw.save() + + if JobLog.objects.filter(task_id=self.request.id).exists(): + JobLog.objects.filter(task_id=self.request.id).update( + status="FAILED", task_result=pw.url + ) + raise e pw.kv.update(**{"status": "completed"}) pw.save() + + if JobLog.objects.filter(task_id=self.request.id).exists(): + JobLog.objects.filter(task_id=self.request.id).update(status="SUCCESS", task_result=pw.url) + + return pw.url + + +@shared_task(bind=True, queue="background") +def identify_missing_rules( + self, + pw_pks: List[int], + rule_package_pk: int, +): + from utilities.misc import PathwayUtils + + rules = Package.objects.get(pk=rule_package_pk).get_applicable_rules() + + rows: List[Any] = [] + header = [ + "Package Name", + "Pathway Name", + "Educt Name", + "Educt SMILES", + "Reaction Name", + "Reaction SMIRKS", + "Triggered Rules", + "Reactant SMARTS", + "Product SMARTS", + "Product Names", + "Product SMILES", + ] + + rows.append(header) + + for pw in Pathway.objects.filter(pk__in=pw_pks): + pu = PathwayUtils(pw) + + missing_rules = pu.find_missing_rules(rules) + + package_name = pw.package.name + pathway_name = pw.name + + for edge_url, rule_chain in missing_rules.items(): + row: List[Any] = [package_name, pathway_name] + edge = Edge.objects.get(url=edge_url) + educts = edge.start_nodes.all() + + for educt in educts: + row.append(educt.default_node_label.name) + row.append(educt.default_node_label.smiles) + + row.append(edge.edge_label.name) + row.append(edge.edge_label.smirks()) + + rule_names = [] + reactant_smarts = [] + product_smarts = [] + + for r in rule_chain: + r = Rule.objects.get(url=r[0]) + rule_names.append(r.name) + + rs = r.reactants_smarts + if isinstance(rs, set): + rs = list(rs) + + ps = r.products_smarts + if isinstance(ps, set): + ps = list(ps) + + reactant_smarts.append(rs) + product_smarts.append(ps) + + row.append(rule_names) + row.append(reactant_smarts) + row.append(product_smarts) + + products = edge.end_nodes.all() + product_names = [] + product_smiles = [] + + for product in products: + product_names.append(product.default_node_label.name) + product_smiles.append(product.default_node_label.smiles) + + row.append(product_names) + row.append(product_smiles) + + rows.append(row) + + buffer = io.StringIO() + + writer = csv.writer(buffer) + writer.writerows(rows) + + buffer.seek(0) + + return buffer.getvalue() diff --git a/epdb/templatetags/envipytags.py b/epdb/templatetags/envipytags.py index c8c92fef..6c250e63 100644 --- a/epdb/templatetags/envipytags.py +++ b/epdb/templatetags/envipytags.py @@ -1,8 +1,21 @@ from django import template +from pydantic import AnyHttpUrl, ValidationError +from pydantic.type_adapter import TypeAdapter register = template.Library() +url_adapter = TypeAdapter(AnyHttpUrl) + @register.filter def classname(obj): return obj.__class__.__name__ + + +@register.filter +def is_url(value): + try: + url_adapter.validate_python(value) + return True + except ValidationError: + return False diff --git a/epdb/urls.py b/epdb/urls.py index 391a2f32..25e18680 100644 --- a/epdb/urls.py +++ b/epdb/urls.py @@ -190,6 +190,7 @@ urlpatterns = [ re_path(r"^indigo/dearomatize$", v.dearomatize, name="indigo_dearomatize"), re_path(r"^indigo/layout$", v.layout, name="indigo_layout"), re_path(r"^depict$", v.depict, name="depict"), + re_path(r"^jobs", v.jobs, name="jobs"), # OAuth Stuff path("o/userinfo/", v.userinfo, name="oauth_userinfo"), ] diff --git a/epdb/views.py b/epdb/views.py index 7f8d2c41..1a2ce23c 100644 --- a/epdb/views.py +++ b/epdb/views.py @@ -47,6 +47,7 @@ from .models import ( ExternalDatabase, ExternalIdentifier, EnzymeLink, + JobLog, ) logger = logging.getLogger(__name__) @@ -236,6 +237,7 @@ def get_base_context(request, for_user=None) -> Dict[str, Any]: "enabled_features": s.FLAGS, "debug": s.DEBUG, "external_databases": ExternalDatabase.get_databases(), + "site_id": s.MATOMO_SITE_ID, }, } @@ -754,8 +756,8 @@ def package_models(request, package_uuid): context["unreviewed_objects"] = unreviewed_model_qs context["model_types"] = { - "ML Relative Reasoning": "ml-relative-reasoning", - "Rule Based Relative Reasoning": "rule-based-relative-reasoning", + "ML Relative Reasoning": "mlrr", + "Rule Based Relative Reasoning": "rbrr", } if s.FLAGS.get("ENVIFORMER", False): @@ -775,69 +777,67 @@ def package_models(request, package_uuid): model_type = request.POST.get("model-type") + # Generic fields for ML and Rule Based + rule_packages = request.POST.getlist("model-rule-packages") + data_packages = request.POST.getlist("model-data-packages") + + # Generic params + params = { + "package": current_package, + "name": name, + "description": description, + "data_packages": [ + PackageManager.get_package_by_url(current_user, p) for p in data_packages + ], + } + if model_type == "enviformer": - threshold = float(request.POST.get(f"{model_type}-threshold", 0.5)) + threshold = float(request.POST.get("model-threshold", 0.5)) + params["threshold"] = threshold - mod = EnviFormer.create(current_package, name, description, threshold) + mod = EnviFormer.create(**params) + elif model_type == "mlrr": + # ML Specific + threshold = float(request.POST.get("model-threshold", 0.5)) + # TODO handle additional fingerprinter + # fingerprinter = request.POST.get("model-fingerprinter") - elif model_type == "ml-relative-reasoning" or model_type == "rule-based-relative-reasoning": - # Generic fields for ML and Rule Based - rule_packages = request.POST.getlist("package-based-relative-reasoning-rule-packages") - data_packages = request.POST.getlist("package-based-relative-reasoning-data-packages") - eval_packages = request.POST.getlist( - "package-based-relative-reasoning-evaluation-packages", [] - ) + params["rule_packages"] = [ + PackageManager.get_package_by_url(current_user, p) for p in rule_packages + ] - # Generic params - params = { - "package": current_package, - "name": name, - "description": description, - "rule_packages": [ - PackageManager.get_package_by_url(current_user, p) for p in rule_packages - ], - "data_packages": [ - PackageManager.get_package_by_url(current_user, p) for p in data_packages - ], - "eval_packages": [ - PackageManager.get_package_by_url(current_user, p) for p in eval_packages - ], - } + # App Domain related parameters + build_ad = request.POST.get("build-app-domain", False) == "on" + num_neighbors = request.POST.get("num-neighbors", 5) + reliability_threshold = request.POST.get("reliability-threshold", 0.5) + local_compatibility_threshold = request.POST.get("local-compatibility-threshold", 0.5) - if model_type == "ml-relative-reasoning": - # ML Specific - threshold = float(request.POST.get(f"{model_type}-threshold", 0.5)) - # TODO handle additional fingerprinter - # fingerprinter = request.POST.get(f"{model_type}-fingerprinter") + params["threshold"] = threshold + # params['fingerprinter'] = fingerprinter + params["build_app_domain"] = build_ad + params["app_domain_num_neighbours"] = num_neighbors + params["app_domain_reliability_threshold"] = reliability_threshold + params["app_domain_local_compatibility_threshold"] = local_compatibility_threshold - # App Domain related parameters - build_ad = request.POST.get("build-app-domain", False) == "on" - num_neighbors = request.POST.get("num-neighbors", 5) - reliability_threshold = request.POST.get("reliability-threshold", 0.5) - local_compatibility_threshold = request.POST.get( - "local-compatibility-threshold", 0.5 - ) + mod = MLRelativeReasoning.create(**params) + elif model_type == "rbrr": + params["rule_packages"] = [ + PackageManager.get_package_by_url(current_user, p) for p in rule_packages + ] - params["threshold"] = threshold - # params['fingerprinter'] = fingerprinter - params["build_app_domain"] = build_ad - params["app_domain_num_neighbours"] = num_neighbors - params["app_domain_reliability_threshold"] = reliability_threshold - params["app_domain_local_compatibility_threshold"] = local_compatibility_threshold - - mod = MLRelativeReasoning.create(**params) - else: - mod = RuleBasedRelativeReasoning.create(**params) - - from .tasks import build_model - - build_model.delay(mod.pk) + mod = RuleBasedRelativeReasoning.create(**params) + elif s.FLAGS.get("PLUGINS", False) and model_type in s.CLASSIFIER_PLUGINS.values(): + pass else: return error( request, "Invalid model type.", f'Model type "{model_type}" is not supported."' ) - return redirect(mod.url) + from .tasks import dispatch, build_model + + dispatch(current_user, build_model, mod.pk) + + return redirect(mod.url) else: return HttpResponseNotAllowed(["GET", "POST"]) @@ -865,6 +865,10 @@ def package_model(request, package_uuid, model_uuid): return JsonResponse({"error": f'"{smiles}" is not a valid SMILES'}, status=400) if classify: + from epdb.tasks import dispatch_eager, predict_simple + + res = dispatch_eager(current_user, predict_simple, current_model.pk, stand_smiles) + pred_res = current_model.predict(stand_smiles) res = [] @@ -909,9 +913,25 @@ def package_model(request, package_uuid, model_uuid): current_model.delete() return redirect(current_package.url + "/model") elif hidden == "evaluate": - from .tasks import evaluate_model + from .tasks import dispatch, evaluate_model + + eval_type = request.POST.get("model-evaluation-type") + + if eval_type not in ["sg", "mg"]: + return error( + request, + "Invalid evaluation type", + f'Evaluation type "{eval_type}" is not supported. Only "sg" and "mg" are supported.', + ) + + multigen = eval_type == "mg" + + eval_packages = request.POST.getlist("model-evaluation-packages") + eval_package_ids = [ + PackageManager.get_package_by_url(current_user, p).id for p in eval_packages + ] + dispatch(current_user, evaluate_model, current_model.pk, multigen, eval_package_ids) - evaluate_model.delay(current_model.pk) return redirect(current_model.url) else: return HttpResponseBadRequest() @@ -1809,9 +1829,9 @@ def package_pathways(request, package_uuid): pw.setting = prediction_setting pw.save() - from .tasks import predict + from .tasks import dispatch, predict - predict.delay(pw.pk, prediction_setting.pk, limit=limit) + dispatch(current_user, predict, pw.pk, prediction_setting.pk, limit=limit) return redirect(pw.url) @@ -1847,6 +1867,25 @@ def package_pathway(request, package_uuid, pathway_uuid): return response + if ( + request.GET.get("identify-missing-rules", False) == "true" + and request.GET.get("rule-package") is not None + ): + from .tasks import dispatch_eager, identify_missing_rules + + rule_package = PackageManager.get_package_by_url( + current_user, request.GET.get("rule-package") + ) + res = dispatch_eager( + current_user, identify_missing_rules, [current_pathway.pk], rule_package.pk + ) + + filename = f"{current_pathway.name.replace(' ', '_')}_{current_pathway.uuid}.csv" + response = HttpResponse(res, content_type="text/csv") + response["Content-Disposition"] = f'attachment; filename="{filename}"' + + return response + # Pathway d3_json() relies on a lot of related objects (Nodes, Structures, Edges, Reaction, Rules, ...) # we will again fetch the current pathway identified by this url, but this time together with nearly all # related objects @@ -1930,10 +1969,16 @@ def package_pathway(request, package_uuid, pathway_uuid): if node_url: n = current_pathway.get_node(node_url) - from .tasks import predict + from .tasks import dispatch, predict + + dispatch( + current_user, + predict, + current_pathway.pk, + current_pathway.setting.pk, + node_pk=n.pk, + ) - # Dont delay? - predict(current_pathway.pk, current_pathway.setting.pk, node_pk=n.pk) return JsonResponse({"success": current_pathway.url}) return HttpResponseBadRequest() @@ -2705,6 +2750,24 @@ def setting(request, setting_uuid): pass +def jobs(request): + current_user = _anonymous_or_real(request) + context = get_base_context(request) + + if request.method == "GET": + context["object_type"] = "joblog" + context["breadcrumbs"] = [ + {"Home": s.SERVER_URL}, + {"Jobs": s.SERVER_URL + "/jobs"}, + ] + if current_user.is_superuser: + context["jobs"] = JobLog.objects.all().order_by("-created") + else: + context["jobs"] = JobLog.objects.filter(user=current_user).order_by("-created") + + return render(request, "collections/joblog.html", context) + + ########### # KETCHER # ########### diff --git a/templates/actions/objects/pathway.html b/templates/actions/objects/pathway.html index 28f74443..785f6213 100644 --- a/templates/actions/objects/pathway.html +++ b/templates/actions/objects/pathway.html @@ -22,6 +22,10 @@ Download Pathway as Image {% if meta.can_edit %} +
  • + + Identify Missing Rules +
  • diff --git a/templates/collections/joblog.html b/templates/collections/joblog.html new file mode 100644 index 00000000..7075e08e --- /dev/null +++ b/templates/collections/joblog.html @@ -0,0 +1,71 @@ +{% extends "framework.html" %} +{% load static %} +{% load envipytags %} +{% block content %} + +
    +
    +
    + Jobs +
    +
    +

    + Job Logs Desc +

    + +
    + +
    +

    + + Jobs + +

    +
    +
    +
    + + + + + + + + + + + {% for job in jobs %} + + + + + + + {% if job.task_result and job.task_result|is_url == True %} + + {% elif job.task_result %} + + {% else %} + + {% endif %} + + {% endfor %} + +
    IDNameStatusQueuedDoneResult
    {{ job.task_id }}{{ job.job_name }}{{ job.status }}{{ job.created }}{{ job.done_at }}Result{{ job.task_result|slice:"40" }}...Empty
    +
    +
    + + + +
    +
    +{% endblock content %} diff --git a/templates/framework.html b/templates/framework.html index b9cdfb48..80c7a6d5 100644 --- a/templates/framework.html +++ b/templates/framework.html @@ -56,7 +56,7 @@ (function () { var u = "//matomo.envipath.com/"; _paq.push(['setTrackerUrl', u + 'matomo.php']); - _paq.push(['setSiteId', '10']); + _paq.push(['setSiteId', '{{ meta.site_id }}']); var d = document, g = d.createElement('script'), s = d.getElementsByTagName('script')[0]; g.async = true; g.src = u + 'matomo.js'; diff --git a/templates/modals/collections/new_model_modal.html b/templates/modals/collections/new_model_modal.html index b58a65ed..b5e903b6 100644 --- a/templates/modals/collections/new_model_modal.html +++ b/templates/modals/collections/new_model_modal.html @@ -18,113 +18,117 @@ prediction. You just need to set a name and the packages you want the object to be based on. There are multiple types of models available. For additional information have a look at our - wiki >> + wiki + >> + + + + + - -
    - - - {% for obj in meta.readable_packages %} - {% if obj.reviewed %} - - {% endif %} + {% if obj.reviewed %} + + {% endif %} {% endfor %} {% for obj in meta.readable_packages %} - {% if not obj.reviewed %} - - {% endif %} + {% if not obj.reviewed %} + + {% endif %} {% endfor %} - - - - -
    - - - - {% if meta.enabled_features.PLUGINS and additional_descriptors %} - - - - {% endif %} - - - -
    - {% if meta.enabled_features.APPLICABILITY_DOMAIN %} - -
    - -
    - - {% endif %}
    - -
    - - + + +
    + + +
    + + +
    + + +
    + + +
    + + +
    + +
    + {% if meta.enabled_features.APPLICABILITY_DOMAIN %} + +
    + +
    + + {% endif %}
    @@ -137,53 +141,47 @@ diff --git a/templates/modals/objects/evaluate_model_modal.html b/templates/modals/objects/evaluate_model_modal.html index a42c68bb..1d4b3801 100644 --- a/templates/modals/objects/evaluate_model_modal.html +++ b/templates/modals/objects/evaluate_model_modal.html @@ -17,10 +17,10 @@ For evaluation, you need to select the packages you want to use. While the model is evaluating, you can use the model for predictions. - - - {% for obj in meta.readable_packages %} {% if obj.reviewed %} @@ -35,7 +35,16 @@ {% endif %} {% endfor %} - + + + + + +
    @@ -50,7 +59,7 @@ $(function () { - $("#relative-reasoning-evaluation-packages").selectpicker(); + $("#model-evaluation-packages").selectpicker(); $('#evaluate_model_form_submit').on('click', function (e) { e.preventDefault(); diff --git a/templates/modals/objects/identify_missing_rules_modal.html b/templates/modals/objects/identify_missing_rules_modal.html new file mode 100644 index 00000000..23f2a953 --- /dev/null +++ b/templates/modals/objects/identify_missing_rules_modal.html @@ -0,0 +1,54 @@ +{% load static %} + + + diff --git a/templates/objects/model.html b/templates/objects/model.html index b64bee39..4cf4acd2 100644 --- a/templates/objects/model.html +++ b/templates/objects/model.html @@ -117,7 +117,7 @@ {% endif %} - {% if model.app_domain %} + {% if model.ready_for_prediction and model.app_domain %}

    diff --git a/templates/objects/pathway.html b/templates/objects/pathway.html index 4e4cc27d..faa38686 100644 --- a/templates/objects/pathway.html +++ b/templates/objects/pathway.html @@ -83,6 +83,7 @@ {% include "modals/objects/add_pathway_edge_modal.html" %} {% include "modals/objects/download_pathway_csv_modal.html" %} {% include "modals/objects/download_pathway_image_modal.html" %} + {% include "modals/objects/identify_missing_rules_modal.html" %} {% include "modals/objects/generic_copy_object_modal.html" %} {% include "modals/objects/edit_pathway_modal.html" %} {% include "modals/objects/generic_set_aliases_modal.html" %} diff --git a/tests/test_enviformer.py b/tests/test_enviformer.py index 28f24126..ff14cdd8 100644 --- a/tests/test_enviformer.py +++ b/tests/test_enviformer.py @@ -3,7 +3,7 @@ from datetime import datetime from tempfile import TemporaryDirectory from django.test import TestCase, tag from epdb.logic import PackageManager -from epdb.models import User, EnviFormer, Package, Setting, Pathway +from epdb.models import User, EnviFormer, Package, Setting from epdb.tasks import predict_simple, predict @@ -48,9 +48,7 @@ class EnviFormerTest(TestCase): mod.build_dataset() mod.build_model() - mod.multigen_eval = True - mod.save() - mod.evaluate_model(n_splits=2) + mod.evaluate_model(True, eval_packages_objs, n_splits=2) mod.predict("CCN(CC)C(=O)C1=CC(=CC=C1)C") diff --git a/tests/test_model.py b/tests/test_model.py index b6c2cce5..bcee3992 100644 --- a/tests/test_model.py +++ b/tests/test_model.py @@ -30,7 +30,6 @@ class ModelTest(TestCase): self.package, rule_package_objs, data_package_objs, - eval_packages_objs, threshold=threshold, name="ECC - BBD - 0.5", description="Created MLRelativeReasoning in Testcase", @@ -38,9 +37,7 @@ class ModelTest(TestCase): mod.build_dataset() mod.build_model() - mod.multigen_eval = True - mod.save() - mod.evaluate_model(n_splits=2) + mod.evaluate_model(True, eval_packages_objs, n_splits=2) results = mod.predict("CCN(CC)C(=O)C1=CC(=CC=C1)C") diff --git a/tests/views/test_model_views.py b/tests/views/test_model_views.py index 558277f5..10cbefe2 100644 --- a/tests/views/test_model_views.py +++ b/tests/views/test_model_views.py @@ -6,7 +6,7 @@ from epdb.logic import UserManager from epdb.models import Package, User -@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models") +@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models", CELERY_TASK_ALWAYS_EAGER=True) class PathwayViewTest(TestCase): fixtures = ["test_fixtures_incl_model.jsonl.gz"] diff --git a/tests/views/test_pathway_views.py b/tests/views/test_pathway_views.py index 9e64e22f..b5fe99fd 100644 --- a/tests/views/test_pathway_views.py +++ b/tests/views/test_pathway_views.py @@ -6,7 +6,7 @@ from epdb.logic import UserManager, PackageManager from epdb.models import Pathway, Edge -@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models") +@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models", CELERY_TASK_ALWAYS_EAGER=True) class PathwayViewTest(TestCase): fixtures = ["test_fixtures_incl_model.jsonl.gz"] diff --git a/utilities/chem.py b/utilities/chem.py index d217ecd0..d7a68d75 100644 --- a/utilities/chem.py +++ b/utilities/chem.py @@ -192,7 +192,7 @@ class FormatConverter(object): return smiles @staticmethod - def standardize(smiles, remove_stereo=False): + def standardize(smiles, remove_stereo=False, canonicalize_tautomers=False): # Taken from https://bitsilla.com/blog/2021/06/standardizing-a-molecule-using-rdkit/ # follows the steps in # https://github.com/greglandrum/RSC_OpenScience_Standardization_202104/blob/main/MolStandardize%20pieces.ipynb @@ -210,19 +210,21 @@ class FormatConverter(object): uncharger = ( rdMolStandardize.Uncharger() ) # annoying, but necessary as no convenience method exists - uncharged_parent_clean_mol = uncharger.uncharge(parent_clean_mol) + res_mol = uncharger.uncharge(parent_clean_mol) # note that no attempt is made at reionization at this step # nor at ionization at some pH (rdkit has no pKa caculator) # the main aim to to represent all molecules from different sources # in a (single) standard way, for use in ML, catalogue, etc. - # te = rdMolStandardize.TautomerEnumerator() # idem - # taut_uncharged_parent_clean_mol = te.Canonicalize(uncharged_parent_clean_mol) if remove_stereo: - Chem.RemoveStereochemistry(uncharged_parent_clean_mol) + Chem.RemoveStereochemistry(res_mol) - return Chem.MolToSmiles(uncharged_parent_clean_mol, kekuleSmiles=True) + if canonicalize_tautomers: + te = rdMolStandardize.TautomerEnumerator() # idem + res_mol = te.Canonicalize(res_mol) + + return Chem.MolToSmiles(res_mol, kekuleSmiles=True) @staticmethod def neutralize_smiles(smiles): @@ -370,6 +372,76 @@ class FormatConverter(object): return parsed_smiles, errors + @staticmethod + def smiles_covered_by( + l_smiles: List[str], + r_smiles: List[str], + standardize: bool = True, + canonicalize_tautomers: bool = True, + ) -> bool: + """ + Check if all SMILES in the left list are covered by (contained in) the right list. + + This function performs a subset check to determine if every chemical structure + represented in l_smiles has a corresponding representation in r_smiles. + + Args: + l_smiles (List[str]): List of SMILES strings to check for coverage. + r_smiles (List[str]): List of SMILES strings that should contain all l_smiles. + standardize (bool, optional): Whether to standardize SMILES before comparison. + Defaults to True. When True, applies FormatConverter.standardize() to + normalize representations for accurate comparison. + canonicalize_tautomers (bool, optional): Whether to canonicalize tautomers + Defaults to False. When True, applies rdMolStandardize.TautomerEnumerator().Canonicalize(res_mol) + to the compounds before comparison. + Returns: + bool: True if all SMILES in l_smiles are found in r_smiles (i.e., l_smiles + is a subset of r_smiles), False otherwise. + + Note: + - Comparison treats lists as sets, ignoring duplicates and order + - Failed standardization attempts are silently ignored (original SMILES used) + - This is a one-directional check: l_smiles ⊆ r_smiles + - For bidirectional equality, both directions must be checked separately + + Example: + >>> FormatConverter.smiles_covered_by(["CCO", "CC"], ["CCO", "CC", "CCC"]) + True + >>> FormatConverter.smiles_covered_by(["CCO", "CCCC"], ["CCO", "CC", "CCC"]) + False + """ + + standardized_l_smiles = [] + + if standardize: + for smi in l_smiles: + try: + smi = FormatConverter.standardize( + smi, canonicalize_tautomers=canonicalize_tautomers + ) + except Exception: + # :shrug: + # logger.debug(f'Standardizing SMILES failed for {smi}') + pass + standardized_l_smiles.append(smi) + else: + standardized_l_smiles = l_smiles + + standardized_r_smiles = [] + if standardize: + for smi in r_smiles: + try: + smi = FormatConverter.standardize(smi) + except Exception: + # :shrug: + # logger.debug(f'Standardizing SMILES failed for {smi}') + pass + standardized_r_smiles.append(smi) + else: + standardized_r_smiles = r_smiles + + return len(set(standardized_l_smiles).difference(set(standardized_r_smiles))) == 0 + class Standardizer(ABC): def __init__(self, name): diff --git a/utilities/misc.py b/utilities/misc.py index 3e4eeb59..0b7222f7 100644 --- a/utilities/misc.py +++ b/utilities/misc.py @@ -9,36 +9,37 @@ from collections import defaultdict from datetime import datetime from enum import Enum from types import NoneType -from typing import Dict, Any, List +from typing import Any, Dict, List from django.db import transaction -from envipy_additional_information import Interval, EnviPyModel -from envipy_additional_information import NAME_MAPPING +from envipy_additional_information import NAME_MAPPING, EnviPyModel, Interval from pydantic import BaseModel, HttpUrl from epdb.models import ( - Package, Compound, CompoundStructure, - SimpleRule, + Edge, + EnviFormer, + EPModel, + ExternalDatabase, + ExternalIdentifier, + License, + MLRelativeReasoning, + Node, + Package, + ParallelRule, + Pathway, + PluginModel, + Reaction, + Rule, + RuleBasedRelativeReasoning, + Scenario, + SequentialRule, SimpleAmbitRule, SimpleRDKitRule, - ParallelRule, - SequentialRule, - Reaction, - Pathway, - Node, - Edge, - Scenario, - EPModel, - MLRelativeReasoning, - RuleBasedRelativeReasoning, - EnviFormer, - PluginModel, - ExternalIdentifier, - ExternalDatabase, - License, + SimpleRule, ) +from utilities.chem import FormatConverter logger = logging.getLogger(__name__) @@ -48,7 +49,7 @@ class HTMLGenerator: @staticmethod def generate_html(additional_information: "EnviPyModel", prefix="") -> str: - from typing import get_origin, get_args, Union + from typing import Union, get_args, get_origin if isinstance(additional_information, type): clz_name = additional_information.__name__ @@ -1171,3 +1172,89 @@ class PackageImporter: url=identifier_data.get("url", ""), is_primary=identifier_data.get("is_primary", False), ) + + +class PathwayUtils: + def __init__(self, pathway: "Pathway"): + self.pathway = pathway + + @staticmethod + def _get_products(smiles: str, rules: List["Rule"]): + educt_rule_products: Dict[str, Dict[str, List[str]]] = defaultdict( + lambda: defaultdict(list) + ) + + for r in rules: + product_sets = r.apply(smiles) + for product_set in product_sets: + for product in product_set: + educt_rule_products[smiles][r.url].append(product) + + return educt_rule_products + + def find_missing_rules(self, rules: List["Rule"]): + print(f"Processing {self.pathway.name}") + # compute products for each node / rule combination in the pathway + educt_rule_products = defaultdict(lambda: defaultdict(list)) + + for node in self.pathway.nodes: + educt_rule_products.update(**self._get_products(node.default_node_label.smiles, rules)) + + # loop through edges and determine reactions that can't be constructed by + # any of the rules or a combination of two rules in a chained fashion + + res: Dict[str, List["Rule"]] = dict() + + for edge in self.pathway.edges: + found = False + reaction = edge.edge_label + + educts = [cs for cs in reaction.educts.all()] + products = [cs.smiles for cs in reaction.products.all()] + rule_chain = [] + + for educt in educts: + educt = educt.smiles + triggered_rules = list(educt_rule_products.get(educt, {}).keys()) + for triggered_rule in triggered_rules: + if rule_products := educt_rule_products[educt][triggered_rule]: + # check if this rule covers the reaction + if FormatConverter.smiles_covered_by( + products, rule_products, standardize=True, canonicalize_tautomers=True + ): + found = True + else: + # Check if another prediction step would cover the reaction + for product in rule_products: + prod_rule_products = self._get_products(product, rules) + prod_triggered_rules = list( + prod_rule_products.get(product, {}).keys() + ) + for prod_triggered_rule in prod_triggered_rules: + if second_step_products := prod_rule_products[product][ + prod_triggered_rule + ]: + if FormatConverter.smiles_covered_by( + products, + second_step_products, + standardize=True, + canonicalize_tautomers=True, + ): + rule_chain.append( + ( + triggered_rule, + Rule.objects.get(url=triggered_rule).name, + ) + ) + rule_chain.append( + ( + prod_triggered_rule, + Rule.objects.get(url=prod_triggered_rule).name, + ) + ) + res[edge.url] = rule_chain + + if not found: + res[edge.url] = rule_chain + + return res