add cons_algrm in pyabpoa

.github/workflows/wheels.yml → .github/workflows/pypi.yml

@@ -16,35 +16,35 @@ env:
   CIBW_SKIP: pp*
 
 jobs:
-  build_wheels:
-    name: Build wheels for ${{ matrix.python }}-${{ matrix.buildplat[1] }}
-    runs-on: ${{ matrix.buildplat[0] }}
-    strategy:
-      # Ensure that a wheel builder finishes even if another fails
-      fail-fast: false
-      matrix:
-        buildplat:
-        - [ubuntu-latest, manylinux_x86_64, auto]
-        - [macos-latest, macosx_x86_64, x86_64]
-        # skip these for now, need more work
-        - [macos-latest, macosx_arm64, arm64]
-        # python: ["cp38", "cp39", "cp310", "cp311", "cp312"]
-        python: ["cp312"]
+  # build_wheels:
+  #   name: Build wheels for ${{ matrix.python }}-${{ matrix.buildplat[1] }}
+  #   runs-on: ${{ matrix.buildplat[0] }}
+  #   strategy:
+  #     # Ensure that a wheel builder finishes even if another fails
+  #     fail-fast: false
+  #     matrix:
+  #       buildplat:
+  #       - [ubuntu-latest, manylinux_x86_64, auto]
+  #       - [macos-latest, macosx_x86_64, x86_64]
+  #       # skip these for now, need more work
+  #       - [macos-latest, macosx_arm64, arm64]
+  #       # python: ["cp38", "cp39", "cp310", "cp311", "cp312"]
+  #       python: ["cp312"]
 
-    steps:
-      - uses: actions/checkout@v3
-        with:
-            submodules: 'true'
+  #   steps:
+  #     - uses: actions/checkout@v3
+  #       with:
+  #           submodules: 'true'
 
-      - name: Build wheels
-        uses: pypa/[email protected]
-        env:
-          CIBW_ARCHS: ${{ matrix.buildplat[2] }}
-          CIBW_BUILD: ${{ matrix.python }}-${{ matrix.buildplat[1] }}
+  #     - name: Build wheels
+  #       uses: pypa/[email protected]
+  #       env:
+  #         CIBW_ARCHS: ${{ matrix.buildplat[2] }}
+  #         CIBW_BUILD: ${{ matrix.python }}-${{ matrix.buildplat[1] }}
 
-      - uses: actions/upload-artifact@v3
-        with:
-          path: ./wheelhouse/*.whl
+  #     - uses: actions/upload-artifact@v3
+  #       with:
+  #         path: ./wheelhouse/*.whl
 
   build_sdist:
     name: Build source distribution
@@ -62,7 +62,8 @@ jobs:
           path: dist/*.tar.gz
 
   upload_pypi:
-    needs: [build_wheels, build_sdist]
+    # needs: [build_wheels, build_sdist]
+    needs: [build_sdist]
     runs-on: ubuntu-latest
     # upload to PyPI on every tag starting with 'v'
     if: github.event_name == 'push' && startsWith(github.ref, 'refs/tags/v')

include/abpoa.h

@@ -36,7 +36,7 @@
 #define ABPOA_OUT_CONS_FQ  5
 
 #define ABPOA_HB 0
-#define ABPOA_MC 1
+#define ABPOA_MF 1
 
 #define ABPOA_NONE_VERBOSE 0
 #define ABPOA_INFO_VERBOSE 1

python/README.md

@@ -63,6 +63,7 @@ This constructs a multiple sequence alignment handler of pyabpoa, it accepts the
 * **gap_ext2**: second gap extension penalty; default: **1**
 * **extra_b**: first adaptive banding paremeter; set as < 0 to disable adaptive banded DP; default: **10**
 * **extra_f**: second adaptive banding paremete; the number of extra bases added on both sites of the band is *b+f\*L*, where *L* is the length of the aligned sequence; default : **0.01**
+* **cons_algrm**: consensus calling algorithm. 'HB': heaviest bunlding, 'MF': most frequent bases; default: **'HB'**
 
 The `msa_aligner` handler provides one method which performs multiple sequence alignment and takes four arguments:
 ```

python/cabpoa.pxd

@@ -35,7 +35,6 @@ cdef extern from "abpoa.h":
     cdef int ABPOA_OUT_CONS_GFA "ABPOA_OUT_CONS_GFA"
 
     cdef int ABPOA_HB "ABPOA_HB"
-    cdef int ABPOA_HC "ABPOA_HC"
     cdef int ABPOA_MF "ABPOA_MF"
 
     ctypedef struct abpoa_res_t:
@@ -63,7 +62,7 @@ cdef extern from "abpoa.h":
         uint8_t use_qv, disable_seeding, progressive_poa
         char *incr_fn
         char *out_pog
-        int align_mode, gap_mode, max_n_cons
+        int align_mode, gap_mode, max_n_cons, cons_algrm
         double min_freq # for diploid data
         int verbose
 

python/pyabpoa.pyx

@@ -90,8 +90,10 @@ cdef class msa_aligner:
     cdef abpoa_para_t abpt
     cdef seq2int_dict, int2seq_dict
 
-    def __cinit__(self, aln_mode='g', is_aa=False, match=2, mismatch=4, score_matrix=b'', gap_open1=4, gap_open2=24, gap_ext1=2, gap_ext2=1,
-            extra_b=10, extra_f=0.01):
+    def __cinit__(self, aln_mode='g', is_aa=False,
+                  match=2, mismatch=4, score_matrix=b'', gap_open1=4, gap_open2=24, gap_ext1=2, gap_ext2=1,
+                  extra_b=10, extra_f=0.01,
+                  cons_algrm='HB'):
         self.ab = abpoa_init()
 
         if aln_mode == 'g':
@@ -132,6 +134,12 @@ cdef class msa_aligner:
         self.abpt.zdrop = -1
         self.abpt.disable_seeding = 1
         self.abpt.progressive_poa = 0
+        if cons_algrm.upper() == 'MF':
+            self.abpt.cons_algrm = ABPOA_MF
+        elif cons_algrm.upper() == 'HB':
+            self.abpt.cons_algrm = ABPOA_HB
+        else:
+            raise Exception('Unknown conseneus calling mode: {}'.format(cons_algrm))
 
         self.seq2int_dict, self.int2seq_dict = set_seq_int_dict(self.abpt.m)
 

src/abpoa.c

@@ -121,7 +121,7 @@ int abpoa_usage(void)
     err_printf("                            - %d: consensus in FASTQ format\n", ABPOA_OUT_CONS_FQ);
     err_printf("    -a --cons-algrm INT     consensus algorithm [%d]\n", ABPOA_HB);
     err_printf("                            - %d: heaviest bundling path in partial order graph\n", ABPOA_HB);
-    err_printf("                            - %d: most frequent bases at each position\n", ABPOA_MC);
+    err_printf("                            - %d: most frequent bases at each position\n", ABPOA_MF);
     err_printf("    -d --maxnum-cons INT    max. number of consensus sequence to generate [1]\n");
     err_printf("    -q --min-freq  FLOAT    min. frequency of each consensus sequence (only effective when -d/--num-cons > 1) [%.2f]\n", MULTIP_MIN_FREQ);
     err_printf("    -g --out-pog    FILE    dump final alignment graph to FILE (.pdf/.png) [Null]\n\n");

src/abpoa.h

@@ -36,7 +36,7 @@
 #define ABPOA_OUT_CONS_FQ  5
 
 #define ABPOA_HB 0
-#define ABPOA_MC 1
+#define ABPOA_MF 1
 
 #define ABPOA_NONE_VERBOSE 0
 #define ABPOA_INFO_VERBOSE 1

src/abpoa_align.c

@@ -143,10 +143,10 @@ abpoa_para_t *abpoa_init_para(void) {
 
 void abpoa_post_set_para(abpoa_para_t *abpt) {
     abpoa_set_gap_mode(abpt);
-    if (abpt->out_msa || abpt->out_gfa || abpt->max_n_cons > 1 || abpt->cons_algrm == ABPOA_MC) {
+    if (abpt->out_msa || abpt->out_gfa || abpt->max_n_cons > 1 || abpt->cons_algrm == ABPOA_MF) {
         abpt->use_read_ids = 1;
         if (abpt->out_msa || abpt->out_gfa) set_65536_table();
-        if (abpt->max_n_cons > 1 || abpt->cons_algrm == ABPOA_MC) set_bit_table16();
+        if (abpt->max_n_cons > 1 || abpt->cons_algrm == ABPOA_MF) set_bit_table16();
     }
     if (abpt->align_mode == ABPOA_LOCAL_MODE) abpt->wb = -1;
     int i;

src/abpoa_graph.c

@@ -287,7 +287,7 @@ void abpoa_topological_sort(abpoa_graph_t *abg, abpoa_para_t *abpt) {
         // fprintf(stderr, "node_n: %d, index_rank_m: %d\n", node_n, abg->index_rank_m);
         abg->index_to_node_id = (int*)_err_realloc(abg->index_to_node_id, abg->index_rank_m * sizeof(int));
         abg->node_id_to_index = (int*)_err_realloc(abg->node_id_to_index, abg->index_rank_m * sizeof(int));
-        if (abpt->out_msa || abpt->max_n_cons > 1 || abpt->cons_algrm == ABPOA_MC) 
+        if (abpt->out_msa || abpt->max_n_cons > 1 || abpt->cons_algrm == ABPOA_MF) 
             abg->node_id_to_msa_rank = (int*)_err_realloc(abg->node_id_to_msa_rank, abg->index_rank_m * sizeof(int));
         if (abpt->wb >= 0) {
             abg->node_id_to_max_pos_left = (int*)_err_realloc(abg->node_id_to_max_pos_left, abg->index_rank_m * sizeof(int));
@@ -699,7 +699,7 @@ void abpoa_reset(abpoa_t *ab, abpoa_para_t *abpt, int qlen) {
         abg->node_m = abg->index_rank_m = node_m;
         abg->index_to_node_id = (int*)_err_realloc(abg->index_to_node_id, node_m * sizeof(int));
         abg->node_id_to_index = (int*)_err_realloc(abg->node_id_to_index, node_m * sizeof(int));
-        if (abpt->out_msa || abpt->max_n_cons > 1 || abpt->cons_algrm == ABPOA_MC) 
+        if (abpt->out_msa || abpt->max_n_cons > 1 || abpt->cons_algrm == ABPOA_MF) 
             abg->node_id_to_msa_rank = (int*)_err_realloc(abg->node_id_to_msa_rank, node_m * sizeof(int));
         if (abpt->wb >= 0) {
             abg->node_id_to_max_pos_left = (int*)_err_realloc(abg->node_id_to_max_pos_left, node_m * sizeof(int));