biopipen 0.31.0__tar.gz → 0.31.2__tar.gz

{biopipen-0.31.0 → biopipen-0.31.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: biopipen
-Version: 0.31.0
+Version: 0.31.2
 Summary: Bioinformatics processes/pipelines that can be run from `pipen run`
 License: MIT
 Author: pwwang

biopipen-0.31.2/biopipen/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.31.2"

{biopipen-0.31.0 → biopipen-0.31.2}/biopipen/scripts/scrna/SeuratMap2Ref.R

@@ -45,6 +45,10 @@ if (is.null(split_by)) {
     future::plan(strategy = "multicore", workers = ncores)
 }
 
+.is_sct <- function(x) {
+    return(Seurat:::IsSCT(assay = x@assays[[DefaultAssay(x)]]))
+}
+
 .expand_dims = function(args, name = "dims") {
     # Expand dims from 30 to 1:30
     if (is.numeric(args[[name]]) && length(args[[name]] == 1)) {
@@ -63,6 +67,8 @@ if (endsWith(ref, ".rds") || endsWith(ref, ".RDS")) {
 } else {
     reference = LoadH5Seurat(ref)
 }
+reference = UpdateSeuratObject(reference)
+reference = UpdateSCTAssays(reference)
 
 # check if refdata exists in the reference
 for (rname in names(mapquery_args$refdata)) {
@@ -84,9 +90,20 @@ for (rname in names(mapquery_args$refdata)) {
     }
 }
 
-if (refnorm == "auto" && DefaultAssay(reference) == "SCT") {
+if (refnorm == "auto" && .is_sct(reference)) {
     refnorm = "SCTransform"
 }
+if (refnorm == "SCTransform") {
+    # Check if the reference is SCTransform'ed
+    if (!.is_sct(reference)) {
+        stop("Reference is not SCTransform'ed")
+    }
+    n_models = length(x = slot(object = reference[[DefaultAssay(reference)]], name = "SCTModel.list"))
+    if (n_models == 0) {
+        stop("Reference doesn't contain SCTModel.")
+    }
+}
+
 log_info("  Normalization method used: {refnorm}")
 if (refnorm == "SCTransform") {
     findtransferanchors_args$normalization.method = "SCT"
@@ -307,14 +324,14 @@ sobj@meta.data = sobj@meta.data %>% mutate(
 Idents(sobj) = ident
 
 # Check if PrepSCTFindMarkers is done
-if (DefaultAssay(sobj) == "SCT") {
+if (.is_sct(sobj) && is.null(sobj@commands$PrepSCTFindMarkers)) {
     log_info("- Running PrepSCTFindMarkers ...")
     sobj <- PrepSCTFindMarkers(sobj)
     # compose a new SeuratCommand to record it to sobj@commands
     commands <- names(pbmc_small@commands)
     scommand <- pbmc_small@commands[[commands[length(commands)]]]
     scommand@time.stamp <- Sys.time()
-    scommand@assay.used <- "SCT"
+    scommand@assay.used <- DefaultAssay(sobj)
     scommand@call.string <- "PrepSCTFindMarkers(object = sobj)"
     scommand@params <- list()
     sobj@commands$PrepSCTFindMarkers <- scommand

{biopipen-0.31.0 → biopipen-0.31.2}/biopipen/scripts/tcr/CDR3AAPhyschem.R

@@ -170,6 +170,8 @@ immdata = readRDS(immdatafile)
 merge_data = function(sam) {
     # Merge the data for one sample from immdata and metadata
     out = immdata$data[[sam]] %>%
+        separate_rows(chain, CDR3.aa, V.name, J.name, sep = ";") %>%
+        filter(chain == "TRB") %>%
         mutate(
             Sample = sam,
             locus = "TCRB",

{biopipen-0.31.0 → biopipen-0.31.2}/biopipen/scripts/tcr/TCRClustering.R

@@ -34,9 +34,17 @@ if (is.null(prefix)) { prefix = immdata$prefix }
 if (is.null(prefix)) { prefix = "" }
 
 get_cdr3aa_df = function() {
-    expand_immdata(immdata, cell_id = "Barcode") %>%
+    out = expand_immdata(immdata, cell_id = "Barcode") %>%
         mutate(Barcode = glue(paste0(prefix, "{Barcode}"))) %>%
-        select(Barcode, CDR3.aa)
+        select(Barcode, CDR3.aa, chain)
+
+    if (on_multi) {
+        out$CDR3.aa = sub(";", "", out$CDR3.aa)
+    } else {
+        out = out %>% separate_rows(chain, CDR3.aa, sep = ";") %>%
+            filter(chain == "TRB")
+    }
+    out %>% select(Barcode, CDR3.aa)
 }
 cdr3aa_df = get_cdr3aa_df()
 
@@ -159,12 +167,14 @@ prepare_input = function() {
     # cdr3col = if (!on_multi) "cdr3" else "CDR3.aa"
     cdr3col = "CDR3.aa"
     for (sample in names(seqdata)) {
-        # cdr3 = bind_rows(cdr3, seqdata[[sample]] %>%
-        #     transmute(aminoAcid=CDR3.aa, vMaxResolved=paste0(V.name, "*01"), Sample=sample))
-        cdr3 = union(
-            cdr3,
-            seqdata[[sample]] %>% pull(cdr3col) %>% unique()
-        )
+        sdata = seqdata[[sample]]
+        if (on_multi) {
+            sdata[[cdr3col]] = sub(";", "", sdata[[cdr3col]])
+        } else {
+            sdata = sdata %>% separate_rows(chain, cdr3col, sep = ";") %>%
+                filter(chain == "TRB")
+        }
+        cdr3 = union(cdr3, unique(sdata[[cdr3col]]))
     }
     cdr3 = unique(cdr3)
 

{biopipen-0.31.0 → biopipen-0.31.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "biopipen"
-version = "0.31.0"
+version = "0.31.2"
 description = "Bioinformatics processes/pipelines that can be run from `pipen run`"
 authors = ["pwwang <pwwang@pwwang.com>"]
 license = "MIT"

{biopipen-0.31.0 → biopipen-0.31.2}/setup.py

@@ -82,7 +82,7 @@ entry_points = \
 
 setup_kwargs = {
     'name': 'biopipen',
-    'version': '0.31.0',
+    'version': '0.31.2',
     'description': 'Bioinformatics processes/pipelines that can be run from `pipen run`',
     'long_description': 'None',
     'author': 'pwwang',

biopipen-0.31.0/biopipen/__init__.py

@@ -1 +0,0 @@
-__version__ = "0.31.0"