BioConductor IRanges coverage counts and identify segments - r

Question

I have a dataset with interval information for a bunch of manufacturing circuits
df <- data.frame(structure(list(circuit = structure(c(2L, 1L, 2L, 1L, 2L, 3L,
1L, 1L, 2L), .Label = c("a", "b", "c"), class = "factor"), start = structure(c(1393621200,
1393627920, 1393628400, 1393631520, 1393650300, 1393646400, 1393656000,
1393668000, 1393666200), class = c("POSIXct", "POSIXt"), tzone = ""),
end = structure(c(1393626600, 1393631519, 1393639200, 1393632000,
1393660500, 1393673400, 1393667999, 1393671600, 1393677000
), class = c("POSIXct", "POSIXt"), tzone = ""), id = structure(1:9, .Label = c("1001",
"1002", "1003", "1004", "1005", "1006", "1007", "1008", "1009"
), class = "factor")), .Names = c("circuit", "start", "end",
"id"), class = "data.frame", row.names = c(NA, -9L)))
Circuit: Identifier for circuit
Start: Time the circuit started running
Finish: Time the circuit stopped running
Id: Unique identifier for the row
I'm able to create a new data set that counts the number of overlapping intervals:
ir <- IRanges(start = as.numeric(df$start), end = as.numeric(df$end), names = df$id)
cov <- coverage(ir)
start_time <- as.POSIXlt(start(cov), origin = "1970-01-01")
end_time <- as.POSIXlt(end(cov), origin = "1970-01-01")
seconds <- runLength(cov)
circuits_running <- runValue(cov)
res <- data.frame(start_time,end_time,seconds,circuits_running)[-1,]
But what I really need is something that looks more like this:
sqldf("select
res.start_time,
res.end_time,
res.seconds,
res.circuits_running,
df.circuit,
df.id
from res left join df on (res.start_time between df.start and df.end)")
The problem is that the sqldf way of using an inequality join is unbearably slow on my full dataset.
How can I get something similar using IRanges alone?
I suspect it has something to do with RangedData but I've not been able to see how to get what I want. Here's what I've tried...
rd <- RangedData(ir, circuit = df$circuit, id = df$id)
coverage(rd) # works but seems to lose the circuit/id info

Answer 1

The coverage can be represented as ranges, dropping the first (the range from 1970 to the first start point)
cov <- coverage(ir)
intervals <- ranges(cov)[-1]
Your query is to find the start of the interval of each circuit, so I narrow the interval to their start coordinate and find overlaps (the first argument is the 'query', the second the 'subject')
olaps <- findOverlaps(narrow(intervals, width(intervals)), ir)
The number of circuits running in a particular interval is
tabulate(queryHits(olaps), queryLength(olaps))
and the actual circuits are
df[subjectHits(olaps), c("circuit", "id")]
The pieces can be knit together as, perhaps
df1 <- cbind(uid=seq_along(intervals),
as.data.frame(intervals),
circuits_running=tabulate(queryHits(olaps), queryLength(olaps)))
df2 <- cbind(uid=queryHits(olaps),
df[subjectHits(olaps), c("circuit", "id")])
merge(df1, df2, by="uid", all=TRUE)
Ranges can have associated with them 'metadata' that is accessible and subset in a coordinated way, so the connection between data.frame and ranges does not have to be so loose and ad hoc. I might instead have
ir <- IRanges(start = as.numeric(df$start), end = as.numeric(df$end))
mcols(ir) <- DataFrame(df)
## ...
mcols(ir[subjectHits(olaps)])
perhaps with as.data.frame() when done with IRanges-land.
It's better to ask your questions about IRanges on the Bioconductor mailing list; no subscription required.