Cobra Types (#237)

Author: JoshLoecker

main/como/rnaseq_gen.py

@@ -168,6 +168,11 @@ async def _build_matrix_results(
     """
     matrix.dropna(subset="ensembl_gene_id", inplace=True)
     conversion = await ensembl_to_gene_id_and_symbol(ids=matrix["ensembl_gene_id"].tolist(), taxon=taxon)
+
+    # If any one column was
+    if any(conversion[col].eq("-").all() for col in conversion.columns):
+        logger.critical(f"Conversion of Ensembl Gene IDs to Entrez IDs and Gene Symbols was empty - is '{taxon}' the correct taxon ID for this data?")
+
     conversion["ensembl_gene_id"] = conversion["ensembl_gene_id"].str.split(",")
     conversion = conversion.explode("ensembl_gene_id")
     conversion.reset_index(inplace=True, drop=True)

main/como/rnaseq_preprocess.py

@@ -367,7 +367,6 @@ async def _write_counts_matrix(
     final_matrix = final_matrix[["ensembl_gene_id", *rna_specific_sample_names]]
 
     output_counts_matrix_filepath.parent.mkdir(parents=True, exist_ok=True)
-    final_matrix.dropna(inplace=True)
     final_matrix.to_csv(output_counts_matrix_filepath, index=False)
     logger.success(f"Wrote gene count matrix for '{rna.value}' RNA at '{output_counts_matrix_filepath}'")
     return final_matrix
@@ -460,6 +459,7 @@ async def _create_config_df(  # noqa: C901
 
         fragment_label = f"{context_name}_{label}_fragment_size.txt"
         frag_paths = [p for p in aux_lookup["fragment"].values() if p.name == fragment_label]
+        print(f"HERE: {prep}")
         if not frag_paths and prep.lower() != RNAType.TRNA.value.lower():
             logger.warning(f"No fragment file for '{label}'; defaulting to 100 bp (needed for zFPKM).")
             mean_frag = 100.0
@@ -682,7 +682,7 @@ async def read_counts(file: Path) -> list[str]:
             )
 
         # Remove NA values from entrez_gene_id dataframe column
-        return conversion["entrez_gene_id"].dropna().tolist()
+        return conversion["entrez_gene_id"].tolist()
 
     logger.info("Fetching gene info - this can take up to 5 minutes depending on the number of genes and your internet connection")
     genes = set(chain.from_iterable(await asyncio.gather(*[read_counts(f) for f in counts_matrix_filepaths])))
@@ -706,10 +706,7 @@ async def read_counts(file: Path) -> list[str]:
         gene_info.at[i, "entrez_gene_id"] = data.get("entrezgene", pd.NA)
         gene_info.at[i, "ensembl_gene_id"] = ensembl_ids
         gene_info.at[i, "size"] = end_pos - start_pos
-
-    gene_info = gene_info[((~gene_info["entrez_gene_id"].isna()) & (~gene_info["ensembl_gene_id"].isna()) & (~gene_info["gene_symbol"].isna()))]
     gene_info.sort_values(by="ensembl_gene_id", inplace=True)
-    gene_info.dropna(inplace=True)
 
     output_filepath.parent.mkdir(parents=True, exist_ok=True)
     gene_info.to_csv(output_filepath, index=False)
@@ -857,10 +854,10 @@ async def rnaseq_preprocess(
 async def _main():
     context_name = "notreatment"
     taxon = 9606
-    como_context_dir = Path("/Users/joshl/Projects/COMO/main/data/COMO_input/notreatment")
-    output_gene_info_filepath = Path("/Users/joshl/Projects/COMO/main/data/results/notreatment/gene_info.csv")
+    como_context_dir = Path("/Users/joshl/Projects/COMO/main/data/COMO_input/naiveB")
+    output_gene_info_filepath = Path("/Users/joshl/Projects/COMO/main/data/results/naiveB/gene_info_fixed.csv")
     output_trna_metadata_filepath = Path("/Users/joshl/Projects/COMO/main/data/config_sheets/trna_config.xlsx")
-    output_trna_count_matrix_filepath = Path("/Users/joshl/Projects/COMO/main/data/results/notreatment/total-rna/totalrna_notreatment.csv")
+    output_trna_count_matrix_filepath = Path("/Users/joshl/Projects/COMO/main/data/results/naiveB/total-rna/totalrna_naiveB.csv")
 
     await rnaseq_preprocess(
         context_name=context_name,

stubs/cobra/__init__.pyi

@@ -0,0 +1,5 @@
+from cobra import flux_analysis as flux_analysis, io as io, medium as medium, sampling as sampling, summary as summary
+from cobra.core import Configuration as Configuration, DictList as DictList, Gene as Gene, Metabolite as Metabolite, Model as Model, Object as Object, Reaction as Reaction, Solution as Solution, Species as Species
+from cobra.util import show_versions as show_versions
+
+__version__: str

stubs/cobra/core/__init__.pyi

@@ -0,0 +1,10 @@
+from cobra.core.configuration import Configuration as Configuration
+from cobra.core.dictlist import DictList as DictList
+from cobra.core.gene import GPR as GPR, Gene as Gene
+from cobra.core.group import Group as Group
+from cobra.core.metabolite import Metabolite as Metabolite
+from cobra.core.model import Model as Model
+from cobra.core.object import Object as Object
+from cobra.core.reaction import Reaction as Reaction
+from cobra.core.solution import Solution as Solution, get_solution as get_solution
+from cobra.core.species import Species as Species

stubs/cobra/core/configuration.pyi

@@ -0,0 +1,32 @@
+import pathlib
+import types
+from numbers import Number
+
+from _typeshed import Incomplete
+
+from .singleton import Singleton
+
+__all__ = ["Configuration"]
+
+
+class Configuration(metaclass=Singleton):
+    tolerance: float
+    lower_bound: Incomplete
+    upper_bound: Incomplete
+    processes: Incomplete
+    max_cache_size: Incomplete
+    cache_expiration: Incomplete
+
+    def __init__(self, **kwargs) -> None: ...
+    @property
+    def solver(self) -> types.ModuleType: ...
+    @solver.setter
+    def solver(self, value) -> None: ...
+    @property
+    def bounds(self) -> tuple[Number | None, Number | None]: ...
+    @bounds.setter
+    def bounds(self, bounds: tuple[Number | None, Number | None]) -> None: ...
+    @property
+    def cache_directory(self) -> pathlib.Path: ...
+    @cache_directory.setter
+    def cache_directory(self, path: pathlib.Path | str) -> None: ...

stubs/cobra/core/dictlist.pyi

@@ -0,0 +1,35 @@
+from .object import Object as Object
+from typing import Any, Callable, Iterable, Iterator, Pattern
+
+class DictList(list):
+    def __init__(self, *args) -> None: ...
+    def has_id(self, id: Object | str) -> bool: ...
+    def get_by_id(self, id: Object | str) -> Object: ...
+    def list_attr(self, attribute: str) -> list: ...
+    def get_by_any(self, iterable: list[str | Object | int]) -> list: ...
+    def query(self, search_function: str | Pattern | Callable, attribute: str | None = None) -> DictList: ...
+    def append(self, entity: Object) -> None: ...
+    def union(self, iterable: Iterable[Object]) -> None: ...
+    def extend(self, iterable: Iterable[Object]) -> None: ...
+    def __sub__(self, other: Iterable[Object]) -> DictList: ...
+    def __isub__(self, other: Iterable[Object]) -> DictList: ...
+    def __add__(self, other: Iterable[Object]) -> DictList: ...
+    def __iadd__(self, other: Iterable[Object]) -> DictList: ...
+    def __reduce__(self) -> tuple[type['DictList'], tuple, dict, Iterator]: ...
+    def index(self, id: str | Object, *args) -> int: ...
+    def __contains__(self, entity: str | Object) -> bool: ...
+    def __copy__(self) -> DictList: ...
+    def insert(self, index: int, entity: Object) -> None: ...
+    def pop(self, *args) -> Object: ...
+    def add(self, x: Object) -> None: ...
+    def remove(self, x: str | Object) -> None: ...
+    def reverse(self) -> None: ...
+    def sort(self, cmp: Callable = None, key: Callable = None, reverse: bool = False) -> None: ...
+    def __getitem__(self, i: int | slice | Iterable | Object | DictList) -> DictList | Object: ...
+    def __setitem__(self, i: slice | int, y: list | Object) -> None: ...
+    def __delitem__(self, index: int) -> None: ...
+    def __getslice__(self, i: int, j: int) -> DictList: ...
+    def __setslice__(self, i: int, j: int, y: list | Object) -> None: ...
+    def __delslice__(self, i: int, j: int) -> None: ...
+    def __getattr__(self, attr: Any) -> Any: ...
+    def __dir__(self) -> list: ...

stubs/cobra/core/formula.pyi

@@ -0,0 +1,15 @@
+from .object import Object as Object
+from _typeshed import Incomplete
+
+element_re: Incomplete
+
+class Formula(Object):
+    formula: Incomplete
+    elements: Incomplete
+    def __init__(self, formula: str | None = None, **kwargs) -> None: ...
+    def __add__(self, other_formula: Formula | str) -> Formula: ...
+    def parse_composition(self) -> None: ...
+    @property
+    def weight(self) -> float: ...
+
+elements_and_molecular_weights: Incomplete

stubs/cobra/core/gene.pyi

@@ -0,0 +1,58 @@
+import sympy.logic.boolalg as spl
+from ..util import resettable as resettable
+from ..util.util import format_long_string as format_long_string
+from .dictlist import DictList as DictList
+from .species import Species as Species
+from _typeshed import Incomplete
+from ast import AST, BoolOp, Expression, Module, Name, NodeTransformer, NodeVisitor
+from sympy import Symbol
+from typing import Iterable
+
+logger: Incomplete
+keywords: Incomplete
+keyword_re: Incomplete
+number_start_re: Incomplete
+replacements: Incomplete
+
+class GPRWalker(NodeVisitor):
+    gene_set: Incomplete
+    def __init__(self, **kwargs) -> None: ...
+    def visit_Name(self, node: Name) -> None: ...
+    def visit_BoolOp(self, node: BoolOp) -> None: ...
+
+class GPRCleaner(NodeTransformer):
+    gene_set: Incomplete
+    def __init__(self, **kwargs) -> None: ...
+    def visit_Name(self, node: Name) -> Name: ...
+    def visit_BinOp(self, node: BoolOp) -> None: ...
+
+def parse_gpr(str_expr: str) -> tuple: ...
+
+class Gene(Species):
+    def __init__(self, id: str = None, name: str = '', functional: bool = True) -> None: ...
+    @property
+    def functional(self) -> bool: ...
+    @functional.setter
+    @resettable
+    def functional(self, value: bool) -> None: ...
+    def knock_out(self) -> None: ...
+
+class GPR(Module):
+    body: list | None
+    def __init__(self, gpr_from: Expression | Module | AST = None, **kwargs) -> None: ...
+    @classmethod
+    def from_string(cls, string_gpr: str) -> GPR: ...
+    @property
+    def genes(self) -> frozenset: ...
+    def update_genes(self) -> None: ...
+    def eval(self, knockouts: DictList | set | str | Iterable = None) -> bool: ...
+    def to_string(self, names: dict = None) -> str: ...
+    def copy(self) -> GPR: ...
+    def __copy__(self) -> GPR: ...
+    def as_symbolic(self, names: dict = None) -> spl.Or | spl.And | Symbol: ...
+    @classmethod
+    def from_symbolic(cls, sympy_gpr: spl.BooleanFunction | Symbol) -> GPR: ...
+    def __eq__(self, other) -> bool: ...
+
+def eval_gpr(expr: Expression | GPR, knockouts: DictList | set) -> bool: ...
+def ast2str(expr: Expression | GPR, level: int = 0, names: dict = None) -> str: ...

stubs/cobra/core/group.pyi

@@ -0,0 +1,16 @@
+from .object import Object as Object
+from _typeshed import Incomplete
+from typing import Iterable
+
+class Group(Object):
+    KIND_TYPES: Incomplete
+    def __init__(self, id: str, name: str = '', members: Iterable | None = None, kind: str | None = None) -> None: ...
+    def __len__(self) -> int: ...
+    @property
+    def members(self) -> set: ...
+    @property
+    def kind(self) -> str: ...
+    @kind.setter
+    def kind(self, kind: str) -> None: ...
+    def add_members(self, new_members: list) -> None: ...
+    def remove_members(self, to_remove: list) -> None: ...

stubs/cobra/core/metabolite.pyi

@@ -0,0 +1,32 @@
+from ..exceptions import OptimizationError as OptimizationError
+from ..util.solver import check_solver_status as check_solver_status
+from ..util.util import format_long_string as format_long_string
+from .formula import elements_and_molecular_weights as elements_and_molecular_weights
+from .species import Species as Species
+from _typeshed import Incomplete
+from cobra.core import Solution as Solution
+from cobra.summary.metabolite_summary import MetaboliteSummary as MetaboliteSummary
+from optlang.interface import Container as Container
+from pandas import DataFrame as DataFrame
+
+element_re: Incomplete
+
+class Metabolite(Species):
+    formula: Incomplete
+    compartment: Incomplete
+    charge: Incomplete
+    def __init__(self, id: str | None = None, formula: str | None = None, name: str | None = '', charge: float | None = None, compartment: str | None = None) -> None: ...
+    @property
+    def constraint(self) -> Container: ...
+    @property
+    def elements(self) -> dict[str, int | float] | None: ...
+    @elements.setter
+    def elements(self, elements_dict: dict[str, int | float]) -> None: ...
+    @property
+    def formula_weight(self) -> int | float: ...
+    @property
+    def y(self) -> float: ...
+    @property
+    def shadow_price(self) -> float: ...
+    def remove_from_model(self, destructive: bool = False) -> None: ...
+    def summary(self, solution: Solution | None = None, fva: float | DataFrame | None = None) -> MetaboliteSummary: ...