Temporal Axis Error

[Dhihram]

No description provided.

[Dhihram]

Hi, I ever utilized the temporal axis for transmission chain in https://epirhandbook.com/en/transmission-chains.html for several months ago. However, today I utilized the same command
plot(
sub,
x_axis = "date_onset",
node_color = "outcome",
col_pal = c(Death = "firebrick", Recover = "green"),
arrow_size = 0.5,
node_size = 13,
label = FALSE,
height = 700,
width = 700
)
and It was error. How to solve this problem. Thank you.

[Dhihram]

Fyi, I was using pacman::p_install_gh("reconhub/epicontacts@timeline") for the temporal axis.

[finlaycampbell]

Hi Dhihram - thanks for pointing that out. I just pushed an update that might fix the issue, could you re-download the package and let me know if it works now?

[Dhihram]

Hi Mr. Campbell, several days ago I have installed reconhub/epicontacts@ttree and its work. Thank you.

[finlaycampbell]

The timeline branch is more updated and is stable (and should hopefully work now) so I would suggest installing that!

[Dhihram]

thank you

[mllstrm]

Hi! I have a similar problem with the x_axis.

This error always comes up when I try to do a temporal axis om the transmission chain:

"Error in fix.by(by.x, x) : 'by' must specify uniquely valid columns"

Example code:

library(outbreaks)
library(epicontacts)

head(mers_korea_2015[[1]])
head(mers_korea_2015[[2]])

x <- make_epicontacts(linelist=mers_korea_2015[[1]],
contacts = mers_korea_2015[[2]],
directed=TRUE)

vis_epicontacts(x, x_axis = "dt_onset", thin = FALSE)

How do I solve this?

[finlaycampbell]

Hi - sorry for the delay in replying to this, I was out of office for a long time.

This is a mysterious error and one I still can't replicate on my end. Could you print the output from the following commands, which might help in tracking down the error (running your example code above first)?

nodes <- data.frame(id = get_id(x, "all"), stringsAsFactors = FALSE)
head(nodes)

head(x$linelist)

nodes <- merge(nodes, x$linelist, by = "id", all = TRUE, sort = FALSE)

Thanks!

[nistara]

Hey Finlay,

This is what I got when I ran your code:

>     id
1 SK_1
2 SK_2
3 SK_3
4 SK_4
5 SK_5
6 SK_6
> >     id age age_class sex        place_infect reporting_ctry
1 SK_1  68     60-69   M         Middle East    South Korea
2 SK_2  63     60-69   F Outside Middle East    South Korea
3 SK_3  76     70-79   M Outside Middle East    South Korea
4 SK_4  46     40-49   F Outside Middle East    South Korea
5 SK_5  50     50-59   M Outside Middle East    South Korea
6 SK_6  71     70-79   M Outside Middle East    South Korea
                                             loc_hosp   dt_onset  dt_report
1 Pyeongtaek St. Mary, Hospital, Pyeongtaek, Gyeonggi 2015-05-11 2015-05-19
2 Pyeongtaek St. Mary, Hospital, Pyeongtaek, Gyeonggi 2015-05-18 2015-05-20
3 Pyeongtaek St. Mary, Hospital, Pyeongtaek, Gyeonggi 2015-05-20 2015-05-20
4 Pyeongtaek St. Mary, Hospital, Pyeongtaek, Gyeonggi 2015-05-25 2015-05-26
5                           365 Yeollin Clinic, Seoul 2015-05-25 2015-05-27
6 Pyeongtaek St. Mary, Hospital, Pyeongtaek, Gyeonggi 2015-05-24 2015-05-28
  week_report dt_start_exp dt_end_exp    dt_diag outcome   dt_death
1     2015_21   2015-04-18 2015-05-04 2015-05-20   Alive       <NA>
2     2015_21   2015-05-15 2015-05-20 2015-05-20   Alive       <NA>
3     2015_21   2015-05-16 2015-05-16 2015-05-21    Dead 2015-06-04
4     2015_22   2015-05-16 2015-05-20 2015-05-26   Alive       <NA>
5     2015_22   2015-05-17 2015-05-17 2015-05-26   Alive       <NA>
6     2015_22   2015-05-15 2015-05-17 2015-05-28    Dead 2015-06-01

I think the error's coming from this line in the vis_epicontacts' code. At least, that's where I came across it when I tried to rerun the function code.

[finlaycampbell]

Hi everyone, I see what the issue is - the x_axis feature is only available on the timeline development branch of epicontacts, but there was still a relic of the old code hidden in the CRAN version that I didn't realise existed, hence the error instead of a warning message telling you to install the correct version. You can install it using remotes::install_github("reconhub/epicontacts@timeline"), let me know if that works.

I've pushed a fix to the main branch to throw a more informative error. Apologies for the confusion, I thought we were all talking about the same version!

[mllstrm]

Hi! I have been away from keybord a couple of days. I installed the new version; remotes::install_github("reconhub/epicontacts@timeline") and it works after that! Thanks!

But I have to use plot-argument to get the x-axis to work, vis_epicontacs does not arrange the network along the x-axis.

The same code again:

`library(outbreaks)
library(epicontacts)

head(mers_korea_2015[[1]])
head(mers_korea_2015[[2]])

x <- make_epicontacts(linelist=mers_korea_2015[[1]],
contacts = mers_korea_2015[[2]],
directed=TRUE)

vis_epicontacts(x, x_axis = "dt_onset", thin = FALSE)

plot(x, x_axis = "dt_onset", thin = FALSE)`

This is the result from plot(x, x_axis = "dt_onset", thin = FALSE) :

And this is the result from vis_epicontacts(x, x_axis = "dt_onset", thin = FALSE)

Thanks again! I wish you all trevlig helg (a realy nice weekend in Swedish)

[finlaycampbell]

Great, glad to hear it works!

Yes, vis_epicontacts is not really supposed to be called directly, as it only plots the free network without the axis. The plot function makes sure the appropriate plotting method is called depending on the function arguments.

Wish you a trevlig helg too!