xcmap_bio 0.0.1__py3-none-any.whl

xcmap_bio/genome/fasta.py

@@ -0,0 +1,52 @@
+from pyfaidx import Fasta
+from typing import Iterator, Tuple
+
+
+class FastaOptions:
+    """
+    高效 FASTA 读取工具类，基于 pyfaidx。
+    支持随机访问区间、获取序列长度、遍历序列。
+    """
+    def __init__(self, fasta_path: str, rebuild_index: bool = False):
+        self.handler = Fasta(fasta_path, rebuild=rebuild_index)
+
+    def fetch_seq(self, seq_id: str, start: int = None, end: int = None):
+        fasta_reader = _FastaReader(self.handler)
+        return fasta_reader.get_seq(seq_id, start, end)
+
+
+class _FastaReader:
+    """
+    高效 FASTA 工具类，基于 pyfaidx。
+    支持随机访问区间、获取序列长度、遍历序列。
+    """
+
+    def __init__(self, handler):
+        self.handler = handler
+
+    def get_seq(self, seq_id: str, start: int = None, end: int = None) -> str:
+        """
+        获取序列区间 [start, end)，0-based 左闭右开。
+        如果 start/end 都为空，返回整条序列。
+        """
+        seq = self.handler[seq_id]
+        if start is None and end is None:
+            return str(seq)
+        return str(seq[start:end])
+
+    def get_length(self, seq_id: str) -> int:
+        """获取序列长度"""
+        return len(self.handler[seq_id])
+
+    def list_seq_ids(self):
+        """返回所有序列 ID"""
+        return list(self.handler.keys())
+
+    def __iter__(self) -> Iterator[Tuple[str, str]]:
+        """遍历所有序列 (seq_id, sequence)"""
+        for seq_id in self.handler.keys():
+            yield seq_id, str(self.handler[seq_id])
+
+    def close(self):
+        """关闭文件句柄"""
+        self.handler.close()

xcmap_bio/genome/gff3.py

@@ -0,0 +1,238 @@
+from functools import lru_cache
+from typing import Optional, Union, List, Tuple
+
+import pandas as pd
+from pydantic import BaseModel
+
+
+class Gff3ColumnName:
+    """
+    序列 ID（染色体、contig 等），必须与参考序列一致, 如chr1
+    """
+    SEQ_ID = "seq_id"
+    """
+    特征来源（预测程序、数据库名等）如Ensembl
+    """
+    SOURCE = "source"
+    """
+    特征类型（使用 SO ontology term）, 如gene, mRNA, exon, CDS
+    """
+    TYPE = "type"
+    """
+    起始位置（1-based, inclusive）
+    """
+    START = "start"
+    """
+    结束位置（1-based, inclusive）
+    """
+    END = "end"
+    """
+    打分值（浮点数，或 . 代表无值）
+    """
+    SCORE = "score"
+    """
+    链信息（+, -, 或 . 未知）
+    """
+    STRAND = "strand"
+    """
+    仅对 CDS 有意义，取值为 0, 1, 2（表示阅读框相对起点偏移），其他特征用 .
+    """
+    PHASE = "phase"
+    """
+    属性字段，key=value 对形式，以 ; 分隔；至少应包含 ID 或 Parent, 如ID=mRNA0001;Parent=gene0001;Name=BRCA1-201
+    """
+    ATTRIBUTES = "attributes"
+
+
+    @classmethod
+    def fetch_values(cls):
+        return [
+            value
+            for name, value in vars(cls).items()
+            if not name.startswith('_') and not callable(value) and name != 'fetch_values'
+        ]
+
+class Gff3Options:
+
+    class PageRequest(BaseModel):
+        """
+        分页查询请求，like query = GFF3Query(df)
+            result, total = (
+            query.filter(seqid=["chr1", "chr2"], feature_type=["gene", "exon"], start=200, end=400)
+            .order(order_by=["seqid", "start"], ascending=[True, False])
+            .paginate(page=1, page_size=2)
+        )
+        """
+        seq_id: Optional[Union[str, List[str], None]] = None
+        type: Optional[Union[str, List[str], None]] = None
+        start: Optional[int] =  None
+        end: Optional[int] =  None
+        order_by: Optional[Union[str, List[str]]] = Gff3ColumnName.START
+        ascending: Optional[Union[bool, List[bool]]] = True
+        page: int = 1
+        size: int = 20
+
+
+    """
+    GFF3 file parser
+    """
+    def __init__(self, gff3_path):
+        self.df = self.read_gff3(gff3_path)
+
+    @lru_cache
+    def read_gff3(self, gff3_path):
+        df =  pd.read_csv(gff3_path,
+                           skiprows=1,
+                           header=None,
+                           sep="\t",
+                           names=Gff3ColumnName.fetch_values()
+                           )
+        def extract_attr(attrs, key):
+            for item in attrs.split(";"):
+                if item.startswith(f"{key}="):
+                    return item.split("=", 1)[1]
+            return None
+        df["ID"] = df[Gff3ColumnName.ATTRIBUTES].apply(lambda x: extract_attr(x, "ID"))
+        df["Parent"] = df[Gff3ColumnName.ATTRIBUTES].apply(lambda x: extract_attr(x, "Parent"))
+        return df
+
+    def fetch_page(self, page_request: PageRequest) -> Tuple[List[dict], int]:
+        """
+        分页查询gff3数据
+        :param page_request:
+        :return:
+        """
+        return _GFF3Query(self.df).filter(
+            seq_id=page_request.seq_id,
+            seq_type=page_request.type,
+            start=page_request.start,
+            end=page_request.end
+        ).order(
+            order_by=page_request.order_by,
+            ascending=page_request.ascending
+        ).paginate(
+            page=page_request.page,
+            size=page_request.size
+        )
+
+    def fetch_by_gene_id(self, gene_id, is_contain_descendants = True):
+        """
+        通过基因ID查询
+        :param is_contain_descendants: 是否包含子节点一下节点
+        :param gene_id:
+        :return:
+        """
+        children_df = self._find_children(gene_id)
+        if children_df.empty:
+            return []
+        results = []
+        if is_contain_descendants:
+            for children_dict in children_df.to_dict(orient="records"):
+                descendants_df = self._find_descendants(children_dict.get("ID"))
+                if children_df.empty:
+                    continue
+                feature_list = descendants_df.to_dict(orient="records")
+                children_dict['feature_list'] = descendants_df.to_dict(orient="records")
+                results.append(children_dict)
+        else:
+            results = children_df.to_dict(orient="records")
+        return results
+
+    @lru_cache
+    def _find_children(self,
+                       parent_id: str,
+                       seq_id: Optional[str] = None,
+                       feature_type: Optional[str] = None) -> pd.DataFrame:
+        df = self.df
+        """查找直接子节点"""
+        result = df[df["Parent"] == parent_id]
+        if seq_id:
+            result = result[result["seq_id"] == seq_id]
+        if feature_type:
+            result = result[result["type"] == feature_type]
+        return result
+
+    @lru_cache
+    def _find_descendants(self,
+                          parent_id: str,
+                          seq_id: Optional[str] = None,
+                          feature_type: Optional[str] = None) -> pd.DataFrame:
+        """查找所有后代（逐层展开，不做递归遍历）"""
+        results = []
+        df = self.df
+        to_visit = [parent_id]
+        while to_visit:
+            children = df[df["Parent"].isin(to_visit)]
+            if children.empty:
+                break
+            results.append(children)
+            # 下一轮查询：用这批 children 的 ID 当作 parent
+            to_visit = children["ID"].tolist()
+
+        if results:
+            result = pd.concat(results, ignore_index=True)
+            if seq_id:
+                result = result[result["seq_id"] == seq_id]
+            if feature_type:
+                result = result[result["type"] == feature_type]
+            return result
+        else:
+            return pd.DataFrame(columns=df.columns)
+
+class _GFF3Query:
+    def __init__(self, df: pd.DataFrame):
+        self.df = df.copy()
+        self.filtered = self.df
+
+    def filter(
+        self,
+        seq_id: Union[str, List[str], None] = None,
+        seq_type: Union[str, List[str], None] = None,
+        start: int = None,
+        end: int = None,
+    ):
+        """条件过滤"""
+        if seq_id is not None:
+            if isinstance(seq_id, list):
+                self.filtered = self.filtered[self.filtered[Gff3ColumnName.SEQ_ID].isin(seq_id)]
+            else:
+                self.filtered = self.filtered[self.filtered[Gff3ColumnName.SEQ_ID] == seq_id]
+
+        if seq_type is not None:
+            if isinstance(seq_type, list):
+                self.filtered = self.filtered[self.filtered[Gff3ColumnName.TYPE].isin(seq_type)]
+            else:
+                self.filtered = self.filtered[self.filtered[Gff3ColumnName.TYPE] == seq_type]
+
+        if start is not None:
+            self.filtered = self.filtered[self.filtered[Gff3ColumnName.START] >= start]
+        if end is not None:
+            self.filtered = self.filtered[self.filtered[Gff3ColumnName.END] <= end]
+
+        return self
+
+    def order(self, order_by: Union[str, List[str]] = Gff3ColumnName.START, ascending: Union[bool, List[bool]] = True):
+        """排序，可支持多字段"""
+        if isinstance(order_by, str):
+            order_by = [order_by]
+        if isinstance(ascending, bool):
+            ascending = [ascending] * len(order_by)
+
+        valid_cols = [col for col in order_by if col in self.filtered.columns]
+        if valid_cols:
+            self.filtered = self.filtered.sort_values(by=valid_cols, ascending=ascending)
+
+        return self
+
+    def paginate(self, page: int = 1, size: int = 20) -> Tuple[List[dict], int]:
+        """分页，并返回 (结果list, 总条数)"""
+        total_count = len(self.filtered)
+        offset = (page - 1) * size
+        page_df = self.filtered.iloc[offset: offset + size]
+        return page_df.to_dict(orient="records"), total_count
+
+
+if __name__ == '__main__':
+    gff3_options = Gff3Options("/Users/zhangyang/Downloads/Ccs.final.gff")
+    result = gff3_options.fetch_by_gene_id('Ccs01G000100', is_contain_descendants=True)
+    print(result)

xcmap_bio-0.0.1.dist-info/METADATA

@@ -0,0 +1,16 @@
+Metadata-Version: 2.3
+Name: xcmap_bio
+Version: 0.0.1
+Summary: Biological tools
+Author: zhangyang
+Author-email: bryanyo1017@126.com
+Requires-Python: >=3.13
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.13
+Requires-Dist: pandas (>=2.3.2,<3.0.0)
+Requires-Dist: pyfaidx (>=0.9.0.3,<0.10.0.0)
+Requires-Dist: pysam (>=0.23.3,<0.24.0)
+Requires-Dist: xcmap (>=0.1.5,<0.2.0)
+Description-Content-Type: text/markdown
+
+

xcmap_bio-0.0.1.dist-info/RECORD

@@ -0,0 +1,9 @@
+xcmap_bio/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+xcmap_bio/genome/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+xcmap_bio/genome/fasta.py,sha256=xeHndi1akC4DVD-vm2aNpRdKUadUXl9pnPOatAIBSLk,1634
+xcmap_bio/genome/gff3.py,sha256=jhq8qgdaryuRTO0Q9pOOAXxUhcE57JEue6_mJK2a4FI,8182
+xcmap_bio/homology/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+xcmap_bio/transcriptone/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+xcmap_bio-0.0.1.dist-info/METADATA,sha256=zMzzQYUcU5PIhC-y8SRnuxnivRueSEoat-L8bk22JoM,458
+xcmap_bio-0.0.1.dist-info/WHEEL,sha256=fGIA9gx4Qxk2KDKeNJCbOEwSrmLtjWCwzBz351GyrPQ,88
+xcmap_bio-0.0.1.dist-info/RECORD,,

xcmap_bio-0.0.1.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: poetry-core 2.1.2
+Root-Is-Purelib: true
+Tag: py3-none-any