biopipen 0.17.6__py3-none-any.whl → 0.18.0__py3-none-any.whl

biopipen/scripts/tcr/TESSA_source/post_analysis.R

@@ -0,0 +1,124 @@
+getmode <- function(v) {
+  uniqv <- unique(v)
+  uniqv[which.max(tabulate(match(v, uniqv)))]
+}
+plot_tessa<-function(tessa_results,folder,labels=NA)
+{
+  library(data.table)
+  # load data
+  meta_dedup=tessa_results$meta_dedup
+  meta=tessa_results$meta
+  master_dist_e=tessa_results$master_dist_e
+  b=tessa_results$b
+  t=tessa_results$t
+  K=tessa_results$K
+  dt=tessa_results$dt
+  de=tessa_results$de
+  ak=tessa_results$ak
+  phi=tessa_results$phi
+  if (!file.exists(folder)) {dir.create(folder)}
+  # are cluster centers more expanded than other members of the clusters?
+  png(paste(folder,"/clone_size.png",sep=""), width=8, height=8, units="in", res=100)
+  tmp=aggregate(meta$barcode,by=list(meta$group_ID),length)
+  meta_dedup$n=NA
+  meta_dedup[tmp$Group.1,"n"]=tmp$x
+  if(!is.na(labels)){
+    tmp2=aggregate(labels,by=list(meta$group_ID),getmode)
+    tmp_keep=data.table(cbind(meta$group_ID,labels))
+    tmp_table=table(tmp_keep)/tmp$x
+    tmp_table=tmp_table[match(tmp2$Group.1,row.names(tmp_table)),]
+    keep=sapply(1:nrow(tmp_table),function(i) tmp_table[i,tmp2$x[i]])
+    meta_dedup[tmp2$Group.1,"labels"]=tmp2$x
+    meta_dedup[tmp2$Group.1[keep<0.6],"labels"]=999
+  }
+  meta_dedup0=meta_dedup[meta_dedup$cluster_number %in% names(phi[phi>2]),]
+  keep=meta_dedup0$group_ID!=meta_dedup0$cluster_number
+  non_center_n=aggregate(meta_dedup0[keep,"n"],by=list(meta_dedup0$cluster_number[keep]),median)
+  if(!is.na(labels)){
+    meta_dedup0$col=c('salmon','blue','grey')[factor(meta_dedup0$labels,levels = c(1,0,999))]
+    #salmon: CD8; blue: CD4; 'grey: mixed center (CD8 and CD4 cells -> same TCRs)
+    #or the center expresses same levels of CD8 and CD4'
+    plot(non_center_n$x,meta_dedup0[non_center_n$Group.1,"n"],xlab="Clone size of non-center TCR groups",
+         ylab="Clone size of center TCR groups",pch=19,col=meta_dedup0[non_center_n$Group.1,"col"])
+    abline(0,1,col="red")
+  }else{
+    plot(non_center_n$x,meta_dedup0[non_center_n$Group.1,"n"],xlab="Clone size of non-center TCR groups",
+       ylab="Clone size of center TCR groups",pch=19)
+    abline(0,1,col="red")
+  }
+  dev.off()
+  mean(non_center_n[,2])
+  mean(meta_dedup0[non_center_n$Group.1,"n"])
+
+  # expression-TCR distance plot
+  png(paste(folder,"/exp_TCR_pair_plot.png",sep=""),width=8,height=12, units="in", res=100)
+  par(mfrow=c(3,2))
+  de_dt_ak=as.data.frame(cbind(de_len=sapply(de,length),dt_len=sapply(dt,length),ak))
+  de_dt_ak=de_dt_ak[order(-de_dt_ak$de_len),]
+  for (k in rownames(de_dt_ak)[1:min(6,K)])
+  {
+    plot(de[[k]],dt[[k]],main=paste("Cluster:",k),pch=19,
+      col=1+as.numeric(as.factor(names(de[[k]]))),xlab="Expression dist.",
+      ylab="TCR dist.")
+    segments(x0=0,y0=0,y1=max(dt[[k]]),x1=ak[k]*max(dt[[k]]),lwd=2,lty=2)
+  }
+  dev.off()
+
+  # density of TCR distances
+  png(paste(folder,"/TCR_dist_density.png",sep=""),width=8,height=6, units="in", res=100)
+  plot(density(as.matrix(dist(t(t/sqrt(b))))^2/2),xlab="TCR distances",ylab="",lwd=3,main="")
+  lines(density(unlist(dt)),lwd=3,col="red")
+  dev.off()
+
+  # exploratory plot at the TCR level
+  png(paste(folder,"/TCR_explore.png",sep=""),width=12,height=12, units="in", res=100)
+  pca_t=prcomp(t(t/sqrt(b)),scale.=F)$x
+  plot(pca_t[,1],pca_t[,2],type="n",xlab="PC1",ylab="PC2")
+  tmp=as.numeric(as.factor(meta_dedup$cluster_number))
+  names(tmp)=meta_dedup$group_ID
+  text(pca_t[,1],pca_t[,2],label=tmp,col=tmp,cex=0.5+0.5*(rownames(pca_t) %in% names(dt)))
+  for (k in 1:K)
+  {
+    for (group in names(dt[[k]]))
+    {
+      segments(x0=pca_t[names(dt)[k],1],y0=pca_t[names(dt)[k],2],
+        x1=pca_t[group,1],y1=pca_t[group,2],col=tmp[group])
+    }
+  }
+  dev.off()
+}
+
+predict_tessa<-function(tessa_results,t_new,cutoff=NA)
+{
+  b=tessa_results$b
+  if(is.na(cutoff)){
+    cutofflist=sapply(tessa_results$dt,function(x) quantile(x,0.5))
+    cutoff=quantile(cutofflist[cutofflist!=0],0.4)
+  }
+  cluster_new=cutree(hclust(dist(t(t_new/sqrt(b)),method='manhattan'),method='single'),h=cutoff)
+}
+
+plot_Tessa_clusters=function(tessa_results,folder){
+  library(igraph)
+  meta=tessa_results$meta_dedup
+  relations=data.frame('source'=meta$group_ID,'target'=meta$cluster_number,stringsAsFactors = F)
+  node_size=setNames(sqrt(as.vector(table(tessa_results$meta$group_ID))),
+                     meta$group_ID)
+  relations = relations[relations[,1]!=relations[,2], ]
+  verticies=data.frame('name'=unique(c(relations$source,relations$target)),stringsAsFactors = F)
+  node_size=node_size[verticies$name]
+  g=graph_from_data_frame(relations,directed=F,vertices = verticies)
+  verticies$group=edge.betweenness.community(g)$membership
+  png(paste(folder,"/TCR_explore_clusters.png",sep=""),width=12,height=12, units="in", res=100)
+  plot.igraph(g,
+       #mark.groups=verticies$group, # group vertices by betweeness indicator (redish blob background)
+       layout=layout.auto,
+       vertex.color = verticies$group, # color vertices by edge betweeness
+       vertex.label=NA, # no vertex label (name)
+       edge.arrow.size=1,
+       edge.arrow.width=1,
+       edge.width=4,
+       edge.lty="solid",
+       vertex.size=node_size)
+  dev.off()
+}

biopipen/scripts/tcr/TESSA_source/real_data.R

@@ -0,0 +1,67 @@
+library(MASS)
+library(LaplacesDemon)
+library(Rtsne)
+
+run_tessa <- function(
+    script_dir,
+    exp_file,
+    contigs_file,
+    cdr3_file,
+    save,
+    is_sampleCluster,
+    fixed_b,
+    xi = 1e+25,
+    g = 0.001,
+    initialize_cluster_factor = 6,
+    max_iter = 1000
+) {
+    source(file.path(script_dir, 'update.R'))
+    source(file.path(script_dir, 'initialization.R'))
+    source(file.path(script_dir, 'MCMC_control.R'))
+    source(file.path(script_dir, 'utility.R'))
+    source(file.path(script_dir, 'post_analysis.R'))
+
+    # the users need to provide these data:
+    # The columns/rows/length of them should be matched up wherever applicable
+    # exp_file: expression data, cells on columns, and genes on rows.
+    #   e can be constructed by PCA or t-SNE, the first row is the first PC, the second row is the second PC, etc
+    # contigs_file: encoded CDR3 values, cells on columns, and embeddings on rows。
+    # cdr3: character vectors of CDR3 sequences
+    # save: a file dir to store tessa results
+    # (optional) sample_id: a column vector of sample categories. If is_sampleCluster=TRUE, users must provide an additional
+    #   column next to the cdr3 column.
+    # (optional) fixed_b: a vector of pre-defined b. The vector must be numerical and has the length of TCR embeddings.
+    exp_data <- read.csv(exp_file, row.names=1, stringsAsFactors=F, check.names = FALSE)
+    n <- ncol(exp_data)
+    tmp <- apply(exp_data, 1, sd)
+    # Run TSNE
+    e <- t(Rtsne(t(exp_data[tmp > quantile(tmp, 0.9), 1:n]), dims = 3)$Y)
+    colnames(e) <- colnames(exp_data)[1:n]
+    contigs_encoded <- read.csv(contigs_file, stringsAsFactors = FALSE)
+    t <- t(contigs_encoded[1:n, -1])
+    meta <- read.csv(cdr3_file, header = TRUE, stringsAsFactors = FALSE)
+    cdr3 <- meta$cdr3
+    if (is_sampleCluster) {
+        sample_id <- meta$sample
+    } else {
+        sample_id <- NULL
+    }
+    if (!is.null(fixed_b)) {
+        b <- read.csv(fixed_b, header = TRUE, stringsAsFactors = F)$b
+    } else {
+        b <- NULL
+    }
+    # the users need to provide these parameters, here are the suggested values
+    hyper_priors <- list(
+        lambda = mean(apply(t, 1, var)),
+        xi = xi,
+        g = g,
+        tau = 100,
+        u = 0.1,
+        v = 0.1,
+        initialize_cluster_factor = initialize_cluster_factor
+    )
+    #save="~/projects/scTCR/data/Tessa_save"
+    # Tessa
+    Tessa(e, cdr3, t, hyper_priors, max_iter, sample_id, save, b)
+}

biopipen/scripts/tcr/TESSA_source/update.R

@@ -0,0 +1,195 @@
+update_t0=function(t,meta_dedup,K,lambda,b,phi,t0)
+{
+  tmp=aggregate(t(t),by=list(meta_dedup$cluster_number),sum)
+  rownames(tmp)=tmp[,1]
+  tmp=t(as.matrix(tmp[,-1]))
+  tmp=tmp[,names(phi),drop=F]
+  for (k in 1:K) # this may be further optimized
+    {t0[[k]]=rnorm(dim(t)[1],tmp[,k]/(b/lambda+phi[k]),sqrt(1/(1/lambda+phi[k]/b)))}
+  t0
+}
+
+update_ak=function(K,dt,de,sigma,a,g,phi,ak)
+{
+  AK=sapply(1:K,function(k) sum(dt[[k]]^2))+1/g
+  BK=sapply(1:K,function(k) sum(dt[[k]]*de[[k]]))+a/g
+  ak_new=rnorm(K,BK/AK,sigma/sqrt(AK))
+  ak[phi>1]=ak_new[phi>1]
+  ak
+}
+
+update_sigma=function(u,v,K,phi,g,ak,a,de,dt,regression_loss)
+{
+  C=u+1+(K+sum(phi))/2
+  D=v+sum((ak-a)^2)/2/g+sum(regression_loss)/2
+  sqrt(rinvgamma(1, shape=C-1, scale=D))
+}
+
+update_a=function(tau,K,g,sigma,ak)
+{
+  E=1/tau+K/(g*sigma^2)
+  D=sum(ak)/(g*sigma^2)
+  a=rtnorm(mu=D/E,sd=sqrt(1/E),a=0,b=Inf)
+  a
+}
+
+update_b=function(u,v,phi,t,t0,meta_dedup,K,dt,de,ak,sigma,regression_loss,b,preset_b)
+{
+  if(!preset_b){
+    # new beta
+    MH_alpha=u+sum(phi)/2
+    MH_beta=v+rowSums((t-unlist(t0[meta_dedup[colnames(t),"cluster_number"]]))^2)/2
+    bnew=sapply(1:length(b),function(q) rinvgamma(1,shape=MH_alpha,scale=MH_beta[q]))
+
+    # new dt
+    dtnew=dt
+    for(k in 1:K)
+    {
+      c=names(phi)[k]
+      group=names(dt[[k]])
+      dtnew[[k]]=colSums((t[,group,drop=F]-t[,c])^2/bnew/2)
+    }
+
+    # F
+    regression_loss_new=sapply(1:K,function(k) sum((de[[k]]-ak[k]*dtnew[[k]])^2))
+    F=sum(regression_loss_new-regression_loss)/2/sigma^2
+
+    updated=0
+    if (runif(1,0,1)<min(1,exp(-F)))
+    {
+      updated=1
+      b=bnew
+      dt=dtnew
+    }
+  }else{
+    b=b;dt=dt;updated=0
+  }
+  return(list(b=b,dt=dt,updated=updated))
+}
+
+calulate_DPprob<-function(phi,xi,group_ID,b,t0_new,ak_new,master_dist_e,sigma,t,ak,t0)
+{
+  t_group_ID=t[,group_ID]
+  part1=c(phi,xi)
+  part2=c(sapply(t0,function(x) sum(-(x-t_group_ID)^2/2/b)),
+    sum(-(t0_new-t_group_ID)^2/2/b))
+  de_test=c(master_dist_e[group_ID,names(phi)],master_dist_e[group_ID,group_ID])
+  dt_test=c(colSums((t_group_ID-t[,names(phi),drop=F])^2/b/2),0)
+  part3=(de_test-dt_test*c(ak,ak_new))^2/(-2*sigma^2)
+  tmp=part2+part3
+  tmp=tmp-max(tmp)
+  part1*exp(tmp)
+}
+
+DP<-function(meta_dedup,meta,t0,dt,de,ak,phi,t,lambda,g,sigma,b,master_dist_e,K,
+  a,xi,mean_t,sample_id)
+{
+  if(!is.null(sample_id)){
+    sample_id_dedup=sample_id[!duplicated(meta$group_ID)]
+    names(sample_id_dedup)=meta_dedup$group_ID
+  }else{
+    sample_id_dedup=NULL
+  }
+  for (group_ID in 1:dim(meta_dedup)[1])
+  {
+    # remove group from old cluster
+    group_to_operate=meta_dedup$group_ID[group_ID]
+    kth_cluster_ind=which(names(ak)==meta_dedup$cluster_number[group_ID])
+
+    if (phi[kth_cluster_ind]==1)
+    {
+      K=K-1
+      phi=phi[-kth_cluster_ind]
+      t0=t0[-kth_cluster_ind]
+      dt=dt[-kth_cluster_ind]
+      de=de[-kth_cluster_ind]
+      ak=ak[-kth_cluster_ind]
+    }else
+    {
+      phi[kth_cluster_ind]=phi[kth_cluster_ind]-1
+    }
+
+    # create a new cluster
+    t0_new=rnorm(n=dim(t)[1],mean_t,sqrt(lambda))
+    ak_new=rnorm(1,a,sigma*sqrt(g))
+
+    # the assignment
+    prob=calulate_DPprob(phi,xi,group_ID,b,t0_new,ak_new,master_dist_e,sigma,t,ak,t0)
+    if(!is.null(sample_id)){
+      sample_id_cluster=sapply(names(prob)[-length(prob)],function(name) strsplit(name,split = ';')[[1]][2])
+      prob2rm=sample_id_cluster!=sample_id_dedup[group_ID]
+      prob2rm=c(prob2rm,FALSE)
+      if(sum(prob[!prob2rm])==0){
+        prob[!prob2rm]=1
+        print(paste('Random selection:',sample_id_dedup[group_ID]))
+      }
+      prob[prob2rm]=0
+    }
+    prob2sc=prob/sum(prob)
+    new_cluster=sample(1:length(prob),1,prob=prob2sc)
+    old_cluster_name=meta_dedup$cluster_number[group_ID]
+
+    # move the group to operate from the old cluster into the new cluster
+    if (new_cluster==length(prob)) # the newly created cluster
+    {
+      meta_dedup$cluster_number[group_ID]=group_to_operate # this CDR3 becomes the center naturally
+      K=K+1
+      # modify the old cluster
+      keep=meta_dedup$cluster_number==old_cluster_name & meta_dedup$group_ID!=group_to_operate
+      go_back=F
+    }else # one of the old clusters
+    {
+      new_cluster_name=names(prob)[new_cluster]
+      meta_dedup$cluster_number[group_ID]=new_cluster_name
+      phi[new_cluster_name]=phi[new_cluster_name]+1
+      # modify the old cluster
+      keep=meta_dedup$cluster_number==old_cluster_name
+      go_back=new_cluster_name==old_cluster_name # the picked cluster is the original cluster
+    }
+
+    # modify the old cluster
+    if (sum(keep)>0 && (!go_back))
+    {
+      old_cluster_name1=find_center(t[,keep,drop=F],b)
+      meta_dedup$cluster_number[keep]=old_cluster_name1
+      which_to_update=which(names(phi)==old_cluster_name)
+      names(phi)[which_to_update]=old_cluster_name1
+      names(t0)=names(ak)=names(dt)=names(de)=names(phi)
+      old_cluster_name=old_cluster_name1
+      tmp=names(de[[old_cluster_name]])
+      groups_to_operate=tmp[tmp!=group_to_operate]
+      de_old_cluster_name=master_dist_e[groups_to_operate,old_cluster_name]
+      dt_old_cluster_name=colSums((t[,groups_to_operate,drop=F]-t[,old_cluster_name])^2/b/2)
+      de[[old_cluster_name]]=named_c(NULL,de_old_cluster_name,groups_to_operate)
+      dt[[old_cluster_name]]=named_c(NULL,dt_old_cluster_name,groups_to_operate)
+    }
+
+    # modify the new cluster
+    if (new_cluster==length(prob)) # new cluster
+    {
+      ak=named_c(ak,ak_new,group_to_operate)
+      phi=named_c(phi,1,group_to_operate)
+      t0[[group_to_operate]]=t0_new
+      tmp=master_dist_e[group_to_operate,group_to_operate]
+      de[[group_to_operate]]=named_c(NULL,tmp,group_to_operate)
+      dt[[group_to_operate]]=named_c(NULL,0,group_to_operate)
+    }else if (!go_back) # one of the old cluster, and also a different one
+    {
+      keep=meta_dedup$cluster_number==new_cluster_name
+      new_cluster_name1=find_center(t[,keep,drop=F],b)
+      meta_dedup$cluster_number[keep]=new_cluster_name1
+      which_to_update=which(names(phi)==new_cluster_name)
+      names(phi)[which_to_update]=new_cluster_name1
+      names(t0)=names(ak)=names(dt)=names(de)=names(phi)
+      new_cluster_name=new_cluster_name1
+      groups_to_operate=c(names(de[[new_cluster_name]]),group_to_operate)
+      de_new_cluster_name=master_dist_e[groups_to_operate,new_cluster_name]
+      dt_new_cluster_name=colSums((t[,groups_to_operate,drop=F]-t[,new_cluster_name])^2/b/2)
+      de[[new_cluster_name]]=named_c(NULL,de_new_cluster_name,groups_to_operate)
+      dt[[new_cluster_name]]=named_c(NULL,dt_new_cluster_name,groups_to_operate)
+    }
+  }
+
+  return(list(cluster_number=meta_dedup$cluster_number,phi=phi,de=de,K=K,
+    dt=dt,ak=ak,t0=t0))
+}

biopipen/scripts/tcr/TESSA_source/utility.R

@@ -0,0 +1,18 @@
+# truncated normal
+rtnorm<-function(mu,sd=1,a=-Inf,b=Inf)
+{
+  #truncated normal distribution sampling function.
+  F<-runif(n=length(mu))
+  Fa<-pnorm((a-mu)/sd,0,sd=1)
+  Fa[a==-Inf] <-0
+  Fb<-pnorm((b-mu)/sd,0,sd=1)
+  Fb[b==Inf]<-1
+  y<-mu+sd*qnorm(F*(Fb-Fa)+Fa)
+  y
+}
+
+named_c<-function(vector,new_element,new_name)
+{
+  names(new_element)=new_name
+  c(vector,new_element)
+}

{biopipen-0.17.6.dist-info → biopipen-0.18.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: biopipen
-Version: 0.17.6
+Version: 0.18.0
 Summary: Bioinformatics processes/pipelines that can be run from `pipen run`
 License: MIT
 Author: pwwang
@@ -13,10 +13,10 @@ Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Provides-Extra: runinfo
-Requires-Dist: datar[pandas] (>=0.13,<0.14)
-Requires-Dist: pipen (>=0.10,<0.11)
-Requires-Dist: pipen-board[report] (>=0.11,<0.12)
-Requires-Dist: pipen-cli-run (>=0.9,<0.10)
-Requires-Dist: pipen-filters (>=0.8,<0.9)
-Requires-Dist: pipen-runinfo (>=0.2,<0.3) ; extra == "runinfo"
-Requires-Dist: pipen-verbose (>=0.7,<0.8)
+Requires-Dist: datar[pandas] (>=0.15.2,<0.16.0)
+Requires-Dist: pipen (>=0.11,<0.12)
+Requires-Dist: pipen-board[report] (>=0.12,<0.13)
+Requires-Dist: pipen-cli-run (>=0.10,<0.11)
+Requires-Dist: pipen-filters (>=0.9,<0.10)
+Requires-Dist: pipen-runinfo (>=0.3,<0.4) ; extra == "runinfo"
+Requires-Dist: pipen-verbose (>=0.8,<0.9)

{biopipen-0.17.6.dist-info → biopipen-0.18.0.dist-info}/RECORD

@@ -1,9 +1,9 @@
-biopipen/__init__.py,sha256=mmrB6n6zH1c3iHQ4iJcecY24GV6KoBQ8Vbb5t5vYe3E,23
+biopipen/__init__.py,sha256=0EHw4xygmgkGSyfwNfEoMlQyN0uHxjHtlSFF79s6120,23
 biopipen/core/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 biopipen/core/config.py,sha256=edK5xnDhM8j27srDzsxubi934NMrglLoKrdcC8qsEPk,1069
 biopipen/core/config.toml,sha256=JALO2S7TfmV3gIRPJ0cLTFWncPXXheQJS3vYQlyX6wQ,1600
 biopipen/core/defaults.py,sha256=yPeehPLk_OYCf71IgRVCWuQRxLAMixDF81Ium0HtPKI,344
-biopipen/core/filters.py,sha256=22VAjUjmkXuVxTD2FgznwKwefVtjhDYUrXR2qsUNa9w,6671
+biopipen/core/filters.py,sha256=NSeEElvRujLZQTV1e4zmQXZb2RjaAvGt1AaRH_xUOTA,7061
 biopipen/core/proc.py,sha256=7TsjBM7EEtMMB-w4jbxV_CSRY8J970gM8320Ga1YeHU,717
 biopipen/core/testing.py,sha256=5vR15kkCjfXM7Bx0HBzabNLtDLAEX4uU94TskCkPni8,1447
 biopipen/ns/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -13,17 +13,17 @@ biopipen/ns/bed.py,sha256=Gu6ctc56NkkuSHoXjJ_fgWXXTCJMtedrtZ8GUX2WEo8,5380
 biopipen/ns/cnv.py,sha256=vq6dZfEOyuVuqg3nP6FQtNmQ-JocpBJMX9IYlZ0OPD0,6803
 biopipen/ns/cnvkit.py,sha256=5mA2Q8-YDs4g1HoxtpB_NWnyZYwEThNr3s3wlubLQrQ,31130
 biopipen/ns/cnvkit_pipeline.py,sha256=_9OusUwS4avpuaqa4TOra74-7pcMbhczTNPrxctrNyM,36769
-biopipen/ns/delim.py,sha256=1rQnqpZDuwD43HWfsXpazxGKkBHsOjv--v2tgDSeQCg,5673
+biopipen/ns/delim.py,sha256=_PTQqxVJUp2t2g6-jPUnjf9C6T11YdYHHkaF3EYJBw0,5683
 biopipen/ns/gene.py,sha256=Q5FzRByfnRITXRNRZR65ApG09FRyiihRC3TcIXxufzE,2228
 biopipen/ns/gsea.py,sha256=EsNRAPYsagaV2KYgr4Jv0KCnZGqayM209v4yOGGTIOI,7423
 biopipen/ns/misc.py,sha256=fzn0pXvdghMkQhu-e3MMapPNMyO6IAJbtTzVU3GbFa0,3246
 biopipen/ns/plot.py,sha256=yguxmErUOH-hOM10JfuI_sXw2p49XF8yGR_gXfbd5yQ,4066
 biopipen/ns/rnaseq.py,sha256=l4vFeRasGhkexopGTM_VfSyIFewOxg-9L5niFzhWUNA,565
-biopipen/ns/scrna.py,sha256=NXI9ljfClE2aTibYXGCQcKi4CHSdFXt7rbu0ASyuhbg,74979
+biopipen/ns/scrna.py,sha256=EIvjkMRtVakruO5X3eoD4j4pWV5TGwf-E48Kav5WB1Q,79087
 biopipen/ns/scrna_basic.py,sha256=os00_hhEaCVRspSCnHvOoItS11OU_aFKgMqbocVyZQU,8538
 biopipen/ns/scrna_metabolic_landscape.py,sha256=AW36EOH-5AJNcMK4j7_LPgJYpyXASH3jj99g_m15SaA,26056
 biopipen/ns/tcgamaf.py,sha256=AFbUJIxiMSvsVY3RcHgjRFuMnNh2DG3Mr5slLNEyz6o,1455
-biopipen/ns/tcr.py,sha256=6C8_B85hOVrW8ntsEEEe4OnZGaIqSL8S1yJPf2DF1ko,57204
+biopipen/ns/tcr.py,sha256=Z02tWLZDSiXMfh9SCKbbwATUN7OcrCB3fOgepAagasY,59922
 biopipen/ns/vcf.py,sha256=cdkKroii0_nl_bSP2cnO09qESUAhHqu6btOiTSKS79Y,15314
 biopipen/ns/web.py,sha256=3zucrDo-IVsSnIvlw-deoScuxqWa6OMTm8Vo-R4E44Q,2224
 biopipen/reports/bam/CNAClinic.svelte,sha256=D4IxQcgDCPQZMbXog-aZP5iJEQTK2N4i0C60e_iXyfs,213
@@ -41,11 +41,11 @@ biopipen/reports/gsea/GSEA.svelte,sha256=lYHf8h8RLx3i-jNCEGu_LM-dbYm9ZJDzyAEadsZ
 biopipen/reports/scrna/CellsDistribution.svelte,sha256=C2Zua-3sXRESttx00cJkQFTUDN5mGbSY0WRyJ5Aoi1Y,1198
 biopipen/reports/scrna/DimPlots.svelte,sha256=ZLbtN0ioevRyEky4jb_DkDGAcpy_jAhaHfFym5ELEPM,479
 biopipen/reports/scrna/GeneExpressionInvistigation.svelte,sha256=nsWHX5VaRSIbUYSFKYcBZaeJIxuUlk5UItA7WPK4clY,793
-biopipen/reports/scrna/MarkersFinder.svelte,sha256=OnakwMU9ZD2tH7nJgG4XxOljAjacJHGj9UBPTHWv754,2770
+biopipen/reports/scrna/MarkersFinder.svelte,sha256=5VKGCNmH0EVfDTINCG6Klr8dxpHgl662EYsS4RNiaKI,3856
 biopipen/reports/scrna/MetaMarkers.svelte,sha256=mPdwV9sTXn6fLTZWqTvFNSn7fdIs-7w4HSerV69o6C0,3709
 biopipen/reports/scrna/RadarPlots.svelte,sha256=QnqQ5J-27KR5apcKmP10fDkoAyO9n3Q2LPEWqDuRxdw,1075
 biopipen/reports/scrna/ScFGSEA.svelte,sha256=uLOlfillgvLbDCO7CzIKZ3HzwO4fZcCcJOpzJRnKwm4,1268
-biopipen/reports/scrna/SeuratClusterStats.svelte,sha256=-EaVQF6A2iRs7FQ6C01AAhdEEnoWMRIWAg-BdO_ukdo,4959
+biopipen/reports/scrna/SeuratClusterStats.svelte,sha256=_vRaffYxpPh_Zdw1O-U7zKkZTUHtrugyhuT7REYaWEk,3433
 biopipen/reports/scrna/SeuratMap2Ref.svelte,sha256=SYuoKJR1PuBf9c7WvUXXjmokVo-JxFLqZqiu9kTtLuY,542
 biopipen/reports/scrna/SeuratPreparing.svelte,sha256=PHeL_prMFTZ-tddhUHG3O6dGUL2ygNNnMEAeHzZVq-k,1182
 biopipen/reports/scrna/TopExpressingGenes.svelte,sha256=lVvlKwMep4woYTiTPTTvlSPOycjxUb44aiEqNzj9u18,1952
@@ -58,6 +58,7 @@ biopipen/reports/tcr/CloneResidency.svelte,sha256=otmBEL1gI3uJjD-ezjwRvklzHHZBXh
 biopipen/reports/tcr/Immunarch.svelte,sha256=IbZgx95-fe4DeoYcxT8o9Z-gqrlLNaar0RbebbFjOfE,6446
 biopipen/reports/tcr/SampleDiversity.svelte,sha256=GrPUpQ6aJPtyQHoFL9ayG6knpW6yZznqwo0ZVZ2tMl4,5524
 biopipen/reports/tcr/TCRClusteringStats.svelte,sha256=gAfrPpo5GtpZKOvA1nEUFZKw5LhtwbEzN2mwmG__Zq4,1865
+biopipen/reports/tcr/TESSA.svelte,sha256=UI7ZeQcLAZXHJQe6pWck9_Hw3ll3YE9L0y_b0cF347Y,2508
 biopipen/reports/tcr/VJUsage.svelte,sha256=c9G2-APGG0l-c6b32aFOimrP_QYD-oELBASN4VabXmk,426
 biopipen/reports/utils/gsea.liq,sha256=5uxNPyIr0xd7nrZePJlIsRyqCPx1uVWso7ehE1F-C4k,3556
 biopipen/reports/utils/misc.liq,sha256=HLK3mkWtIMQqBtM2IIRFUdKKTcY8cvBtyHJB9DbWBdw,1653
@@ -93,7 +94,7 @@ biopipen/scripts/cnvkit/CNVkitScatter.py,sha256=7DhTiXPHEHbdXn0VFcDOR-wTP6sks08N
 biopipen/scripts/cnvkit/CNVkitSegment.py,sha256=q5iGAjY6-yIehPcJpi3hX6EuGre0YgWTPkG_d5LEV48,1629
 biopipen/scripts/cnvkit/guess_baits.py,sha256=7OCMtSMHIJWWZv9qEYVXnB0N4hU_JaGEesKdkr6tvJc,10586
 biopipen/scripts/delim/RowsBinder.R,sha256=yp960u7Ui_jFCL8WDvODa-0vhJvyLo64ll35PzXYUbI,1444
-biopipen/scripts/delim/SampleInfo.R,sha256=e4LSg7_KlNGTZzmb7YFqgSo23uT_uezbllVvNHlRfIs,5096
+biopipen/scripts/delim/SampleInfo.R,sha256=abO6oI2zBzJdKfyL4CvRrQQWkdLD0NYCnhnhYqrosmc,5253
 biopipen/scripts/gene/GeneNameConversion.py,sha256=2RveardTsLv2K1XSj3G0ERYLiln9bcR74bjkRdKcChc,1880
 biopipen/scripts/gsea/Enrichr.R,sha256=tr4vInlVIeiGXumh22ARuTQmy0-Qq869RiX7d7ERqCg,661
 biopipen/scripts/gsea/FGSEA.R,sha256=RLqDgrqnYEacHfzEEuZ3d29lxNqWehigOnGuu248SRg,1483
@@ -116,14 +117,17 @@ biopipen/scripts/scrna/ExprImpution-rmagic.R,sha256=yYnkyVfqIaNynsbaZZLGS6DrAJ_X
 biopipen/scripts/scrna/ExprImpution-scimpute.R,sha256=mg40qCUW7-nP5oHPvARq7dmtoahM0GRFWXQpum0BXVk,1082
 biopipen/scripts/scrna/ExprImpution.R,sha256=7768ezrr59xUZDXq8lO9jj2XhnkSsx-xxBmOD9_DO7c,313
 biopipen/scripts/scrna/GeneExpressionInvistigation.R,sha256=FI5MWic3xRml2DN7ONcyT7pbceOnL30Zd4nBHRZRFNQ,3800
-biopipen/scripts/scrna/MarkersFinder.R,sha256=E99kJOoDV0PvDZgbST3y3Sny3behrWr3GFnMGSAUFF8,8027
+biopipen/scripts/scrna/MarkersFinder.R,sha256=eMLIpJxrdCUNDGmdLL90rNSfKyn6hwxmBvUC4A1S_SE,10089
 biopipen/scripts/scrna/MetaMarkers.R,sha256=d0fPCkD12F9GUes6UkxvMe0e9YqWuV64uRWAcvzJe8s,8331
 biopipen/scripts/scrna/ModuleScoreCalculator.R,sha256=0mLGoTvJRpTbCnmuYbYKqZnP3ZdJQkTn6getJddBKRs,2495
 biopipen/scripts/scrna/RadarPlots.R,sha256=5J1G4c1qfjyI2OTbyDj-XHV8mmh8KWkkuLbO-BNURao,6955
 biopipen/scripts/scrna/SCImpute.R,sha256=dSJOHhmJ3x_72LBRXT72dbCti5oiB85CJ-OjWtqONbk,2958
 biopipen/scripts/scrna/ScFGSEA.R,sha256=puEc9HLzPuHg8DFY6b0Kie3Iyra0H1D4h0r4VX7iIis,4953
-biopipen/scripts/scrna/SeuratClusterStats.R,sha256=HogtO4sVYtH22qMWaQhUDMGBWC3Eo7QU03YGmRzRog0,15740
-biopipen/scripts/scrna/SeuratClustering.R,sha256=H0ix8fJQ-G5i8KLSSK3RJWcKzRCcaHcXRp8LEoXtSiY,6700
+biopipen/scripts/scrna/SeuratClusterStats-dimplots.R,sha256=DKSs32gv1fxpdeCOrrys5n_FnEKEaQniEoKDhrXTvKE,1197
+biopipen/scripts/scrna/SeuratClusterStats-features.R,sha256=vsDys4zTeMfkCYlUnxGt8kO0MuMPozCWu5FjORizY7Y,7438
+biopipen/scripts/scrna/SeuratClusterStats-stats.R,sha256=IKEGW5mjXqpIKbfSLmHm4az2BJqxzM8-Bklnw3ZqhK8,3636
+biopipen/scripts/scrna/SeuratClusterStats.R,sha256=Xst0PIyqrk9QT76hjHW8tK0BZeUHiSoHzGg4l86_i3M,556
+biopipen/scripts/scrna/SeuratClustering.R,sha256=Q_QGWPay3mdn3sy42Ym6kDNnowqctOfcXBSwVUIM-OM,7476
 biopipen/scripts/scrna/SeuratFilter.R,sha256=BrYK0MLdaTtQvInMaQsmOt7oH_hlks0M1zykkJtg2lM,509
 biopipen/scripts/scrna/SeuratLoading.R,sha256=ekWKnHIqtQb3kHVQiVymAHXXqiUxs6KKefjZKjaykmk,900
 biopipen/scripts/scrna/SeuratMap2Ref.R,sha256=TkDSDc5do6BOtkAq3fS5HpjyqWUwsu8YqRC5in62oz8,3750
@@ -161,11 +165,22 @@ biopipen/scripts/tcr/Immunarch-tracking.R,sha256=Zkzh6-9GZS8aKBRyZAegfoiVNUvKu0z
 biopipen/scripts/tcr/Immunarch.R,sha256=Cb7ZkA2-vyqGg4MNQ7C-yFlAXF_kNUo3Q7oak0uNcyo,1214
 biopipen/scripts/tcr/Immunarch2VDJtools.R,sha256=QB9ILGbnsfoWaRANK6ceb14wpSWy8F1V1EdEmfIqiks,706
 biopipen/scripts/tcr/ImmunarchFilter.R,sha256=o25O36FwH_0w6F8DFQ0SfpcwDzlzaGefXqr9ESrvb4k,3974
-biopipen/scripts/tcr/ImmunarchLoading.R,sha256=HS-80YtWB7MYbc8G7pjQB2Ug7X14PHWkCl40JVorg1A,5824
+biopipen/scripts/tcr/ImmunarchLoading.R,sha256=vg3Pg_5JvcOiT3gOrWn98EpjGi8r7SYh52hhvfSyyzk,5836
 biopipen/scripts/tcr/ImmunarchSplitIdents.R,sha256=FGCeGV0uSmFU91lKkldUAeV4A2m3hHw5X4GNi8ffGzI,1873
 biopipen/scripts/tcr/SampleDiversity.R,sha256=jQ1OU3b8vswD8tZhLt3fkcqJKrl2bhQX0giHM2rXz3Y,2643
 biopipen/scripts/tcr/TCRClustering.R,sha256=yfIiCMQuywjoJnAXwRJjlJsoYIA8swUMKIt_AsGvHQY,8566
 biopipen/scripts/tcr/TCRClusteringStats.R,sha256=_NoMX_c22WIYZlVXkJjs7c3YsmEqhWJJl6cXMFa_Srw,8297
+biopipen/scripts/tcr/TESSA.R,sha256=9WuLf3Di4PKb6zclPy7FW1EBITB2He-1vTMRy2vhQI8,6317
+biopipen/scripts/tcr/TESSA_source/Atchley_factors.csv,sha256=SumqDOqP67P54uM7Cuc5_O_rySTWcGo7eX3psMSPX9s,763
+biopipen/scripts/tcr/TESSA_source/BriseisEncoder.py,sha256=z4_Q_6StymffuUGGjHP1-B3aTsXtamKao5Q1-Kg9has,6831
+biopipen/scripts/tcr/TESSA_source/MCMC_control.R,sha256=93Nnz0IG8KfFnVscZDvmBp1qccZoSoG_jIVpOWBQLHE,2911
+biopipen/scripts/tcr/TESSA_source/TrainedEncoder.h5,sha256=U-z4jcmeDXnJQwgmXoIjeJvCj_R8f5wHsesDKYBaF98,457824
+biopipen/scripts/tcr/TESSA_source/fixed_b.csv,sha256=tJJ2hUt0kJLZOxTH_vo_vd_CSbHa_W_JGa3jcOWrczY,570
+biopipen/scripts/tcr/TESSA_source/initialization.R,sha256=TYPciAvWrrOVByfxdsqouEk8k9vgB45oW1jQFfUGLjM,3900
+biopipen/scripts/tcr/TESSA_source/post_analysis.R,sha256=6RX7gdqd1N7lQVfwJBgjfEZEwi1Q4r15mpAthsLSSJ8,5090
+biopipen/scripts/tcr/TESSA_source/real_data.R,sha256=tg3BbiTpRVQWBRZStnzkC2jI1PCmxUA0fjWK09J6grw,2448
+biopipen/scripts/tcr/TESSA_source/update.R,sha256=kVrf6zgIkhhiQ2O55XGmxm_DGBKSpcShFgkyPNQIet0,6661
+biopipen/scripts/tcr/TESSA_source/utility.R,sha256=6qbkMV7yp4bgQe718QiASQUAgOE2euiLgAXOeSqrvHQ,374
 biopipen/scripts/tcr/VJUsage.R,sha256=LjHEbAHW3WriCYiM9-T6Esd4jc6pnoiSxBKTN_YA490,437
 biopipen/scripts/tcr/vdjtools-patch.sh,sha256=rL5qp2S18CrpqduKkeR1HVmmuWhCVJOMKJXvWiKdYIc,566
 biopipen/scripts/vcf/TruvariBench.sh,sha256=80yLQ73OzSgsJ4ltzgpcWxYvvX1hFnCG8YSBhhhRQ9Y,765
@@ -195,7 +210,7 @@ biopipen/utils/plot.R,sha256=pzl37PomNeUZPxohHZ2w93j3Fc4T0Qrc62FF-9MTKdw,4417
 biopipen/utils/reference.py,sha256=6bPSwQa-GiDfr7xLR9a5T64Ey40y24yn3QfQ5wDFZkU,4420
 biopipen/utils/rnaseq.R,sha256=Ro2B2dG-Z2oVaT5tkwp9RHBz4dp_RF-JcizlM5GYXFs,1298
 biopipen/utils/vcf.py,sha256=ajXs0M_QghEctlvUlSRjWQIABVF02wPdYd-0LP4mIsU,9377
-biopipen-0.17.6.dist-info/METADATA,sha256=cppV1wETU_aHmQNPFqvQrfBeY1BRtvGkYrf7L4GkJo0,863
-biopipen-0.17.6.dist-info/WHEEL,sha256=d2fvjOD7sXsVzChCqf0Ty0JbHKBaLYwDbGQDwQTnJ50,88
-biopipen-0.17.6.dist-info/entry_points.txt,sha256=sfI6oDEEuMvAg0KNujE9uu-c29y7IwQQA1_A2sUjPhc,527
-biopipen-0.17.6.dist-info/RECORD,,
+biopipen-0.18.0.dist-info/METADATA,sha256=KctoN8cx0TfZIBSFiXSDNvB64FkCD14Z_07zamsbG1w,869
+biopipen-0.18.0.dist-info/WHEEL,sha256=d2fvjOD7sXsVzChCqf0Ty0JbHKBaLYwDbGQDwQTnJ50,88
+biopipen-0.18.0.dist-info/entry_points.txt,sha256=sfI6oDEEuMvAg0KNujE9uu-c29y7IwQQA1_A2sUjPhc,527
+biopipen-0.18.0.dist-info/RECORD,,