error in AStrap(）

[DXXDR]

result <- AStrap(alignment,trSequence,rice_RFmodel, identity = 0.7, coverage = 0.7, bias = 0, ASlength = 0)

Error in if ((Q.dis <= bias) && (S.dis > 1)) { : missing value where TRUE/FALSE needed

[nikou123456]

Hi,

Thanks for contacting. Maybe your input file is wrong. Could you provide your input file to me? Have you tried with example data? Here is the raw code,

AStrap <- function(cluster.alignment, transcriptSeq, trainModel, identity = 0.7, coverage = 0.7, bias = 0, ASlength = 0) {
  bias <- bias + 1
  ASevent <- data.frame(Name = "", Target = 1, Qname = "", Sname = "", Qstart = 1, Qend = 1, Sstart = 1, Send = 1)
  ASevent$Name <- as.character(ASevent$Name)
  ASevent$Target <- as.numeric(ASevent$Target)
  ASevent$Qname <- as.character(ASevent$Qname)
  ASevent$Sname <- as.character(ASevent$Sname)
  ASevent$Qstart <- as.numeric(ASevent$Qstart)
  ASevent$Qend <- as.numeric(ASevent$Qend)
  ASevent$Sstart <- as.numeric(ASevent$Sstart)
  ASevent$Send <- as.numeric(ASevent$Send)
  as.align <- cluster.alignment[which(cluster.alignment$Alinum > 1), ]
  index.need <- intersect(which(as.align$identity >= identity * 100), which(as.align$Coverage >= coverage * 100))
  as.align <- as.align[index.need, ]
  as.align$Sali <- str_extract_all(as.align$Salign, "\\d+")
  as.align$Qali <- str_extract_all(as.align$Qalign, "\\d+")
  for (i in 1:nrow(as.align)) {
    S.align <- as.numeric(as.align$Sali[[i]])
    Q.align <- as.numeric(as.align$Qali[[i]])
    align.num <- as.numeric(as.align$Alinum[i])
    for (j in 1:(align.num - 1)) {
      S.dis <- S.align[2 * j + 1] - S.align[2 * j]
      Q.dis <- Q.align[2 * j + 1] - Q.align[2 * j]
      if ((Q.dis <= bias) && (S.dis > 1)) {
        TEMP <- data.frame(Name = as.align$NAME[i], Target = 1,
                           Qname = as.align$Qid[i], Sname = as.align$Sid[i],
                           Qstart = Q.align[2 * j], Qend = Q.align[2 * j + 1],
                           Sstart = S.align[2 * j], Send = S.align[2 * j + 1])
        ASevent <- rbind(ASevent, TEMP)

      } else if ((S.dis <= bias) && (Q.dis > 1)) {
        TEMP <- data.frame(Name = as.align$NAME[i], Target = 0,
                           Qname = as.align$Sid[i], Sname = as.align$Qid[i],
                           Qstart = S.align[2 * j],Qend = S.align[2 * j + 1],
                           Sstart = Q.align[2 * j], Send = Q.align[2 * j + 1])
        ASevent <- rbind(ASevent, TEMP)

      }
    }
  }
  ASevent <- ASevent[-1, ]
  rew.ASevent <- cbind(ASevent, as.align[match(as.character(ASevent$Name), as.character(as.align$NAME)), ])
  rew.ASevent <- extract_IsoSeq_tr(rew.ASevent, transcriptSeq)
  row.names(rew.ASevent) <- c(1:nrow(rew.ASevent))
  testfeature <- getFeature(rew.ASevent)
  pred <- predict(trainModel, testfeature)
  rew.ASevent$spliceSeq <- paste0(subseq(rew.ASevent$seq, start = 1, width = 2), "-", subseq(rew.ASevent$seq, end = rew.ASevent$length, width = 2))
  rew.ASevent$Predict <- as.character(pred)
  ASevent <- data.frame(Qid = rew.ASevent$Qname, Qlength = rew.ASevent$QseqLen,
                        Sid = rew.ASevent$Sname, Slength = rew.ASevent$SseqLen,
                        Clusterid = rew.ASevent$QClusterID,CluterSeqNum = rew.ASevent$QseqNum,
                        Qstart = rew.ASevent$Qstart, Qend = rew.ASevent$Qend,
                        Sstart = rew.ASevent$Sstart, Send = rew.ASevent$Send,
                        identity = rew.ASevent$identity, coverage = rew.ASevent$Coverage,
                        ASlength = rew.ASevent$length,Qalign = rew.ASevent$Qalign,
                        Salign =rew.ASevent$Salign,
                        prediction = rew.ASevent$Predict, spliceSeq = rew.ASevent$spliceSeq)
  index <- which(rew.ASevent$Target == 0)
  ASevent$Qlength[index] <- rew.ASevent$SseqLen[index]
  ASevent$Slength[index] <- rew.ASevent$QseqLen[index]
  ASevent$Qalign[index] = rew.ASevent$Qalign[index]
  ASevent$Salign[index] = rew.ASevent$Salign[index]
  ASevent <- ASevent[ASevent$ASlength > ASlength, ]
  return(list(ASevent = ASevent, feature = testfeature, predict = pred))
}

[WangYuPeng1]

I meet the same error, Could you help me?

[nikou123456]

I meet the same error, Could you help me?

Sure, I will sent the raw code to you by Email. You can install this package locally

[nikou123456]

Hi Yupeng,

Thank you for you contacting. Have you successfully installed the package? I have sent the raw code for AStrap to you, you can install this package locally. The directory Astrap/data are example data, you can use these data to test AStrap first. If you still have the same problem as follow:
result <- AStrap(alignment,trSequence,rice_RFmodel, identity = 0.7, coverage = 0.7, bias = 0, ASlength = 0)Error in if ((Q.dis <= bias) && (S.dis > 1)) { : missing value where TRUE/FALSE needed

The value bias means the maximum number of mismatches in pairwise sequence alignment (default:0). You can set this value more larger to see whether it works or not. If it still don't work, may be there is something wrong with you input file.

You alse can find the radw code from this link https://github.com/nikou123456/Supplement.

If you have questions, please let me know.