Notes

This figure contains data from different experiments that were used demonstrate the construction of untargeted and orthogonal double mutants with ORBIT. Data figures were made in R notebooks and exported as pdfs. Cosmetic improvements were made in Adobe Illustrator. Note that Fig. 5A, and 5D-E were made in Adobe Illustrator.

Setup packages and plotting for the notebook:

# Check packages
source("../tools/package_setup.R")
# Load packages
library(tidyverse)
library(cowplot)
library(kableExtra)
# Code display options
knitr::opts_chunk$set(tidy.opts=list(width.cutoff=60),tidy=FALSE, echo = TRUE, message=FALSE, warning=FALSE, fig.align="center", fig.retina = 2)
# Load plotting tools
source("../tools/plotting_tools.R")
#Modify the plot theme
theme_set(theme_notebook())

Fig. 5B - Orthogonal heatmap

Read in the data.

df_heatmap <- read_csv("../../data/low_throughput_experiments/2022_10_14_chlor_orth_2_data.csv") %>% mutate(eff = Chlor_count / LB_count)

df_heatmap %>% kable() %>% kable_styling() %>% scroll_box(height = '250px')

condition	attB	attP	match	LB_count	Chlor_count	eff
attB1_attP1	B1	P1	TRUE	2100000	2900	0.0013810
attB1_attP2	B1	P2	FALSE	2400000	220	0.0000917
attB1_attP3	B1	P3	FALSE	2200000	550	0.0002500
attB1_attP4	B1	P4	FALSE	1200000	3100	0.0025833
attB1_attP5	B1	P5	FALSE	1700000	280	0.0001647
attB1_attP6	B1	P6	FALSE	1700000	860	0.0005059
attB2_attP1	B2	P1	FALSE	1300000	60	0.0000462
attB2_attP2	B2	P2	TRUE	1400000	20000	0.0142857
attB2_attP3	B2	P3	FALSE	1600000	390	0.0002438
attB2_attP4	B2	P4	FALSE	1600000	4200	0.0026250
attB2_attP5	B2	P5	FALSE	1400000	310	0.0002214
attB2_attP6	B2	P6	FALSE	1200000	1400	0.0011667
attB3_attP1	B3	P1	FALSE	1400000	56	0.0000400
attB3_attP2	B3	P2	FALSE	1200000	370	0.0003083
attB3_attP3	B3	P3	TRUE	1600000	7400	0.0046250
attB3_attP4	B3	P4	FALSE	1830000	8400	0.0045902
attB3_attP5	B3	P5	FALSE	1300000	1190	0.0009154
attB3_attP6	B3	P6	FALSE	2400000	680	0.0002833
attB4_attP1	B4	P1	FALSE	1420000	63	0.0000444
attB4_attP2	B4	P2	FALSE	1900000	260	0.0001368
attB4_attP3	B4	P3	FALSE	1200000	610	0.0005083
attB4_attP4	B4	P4	TRUE	1600000	9900	0.0061875
attB4_attP5	B4	P5	FALSE	1900000	400	0.0002105
attB4_attP6	B4	P6	FALSE	1500000	1320	0.0008800
attB5_attP1	B5	P1	FALSE	1200000	53	0.0000442
attB5_attP2	B5	P2	FALSE	2000000	440	0.0002200
attB5_attP3	B5	P3	FALSE	1800000	1070	0.0005944
attB5_attP4	B5	P4	FALSE	1700000	1500	0.0008824
attB5_attP5	B5	P5	TRUE	1500000	8500	0.0056667
attB5_attP6	B5	P6	FALSE	1100000	1160	0.0010545
attB6_attP1	B6	P1	FALSE	1330000	88	0.0000662
attB6_attP2	B6	P2	FALSE	1700000	370	0.0002176
attB6_attP3	B6	P3	FALSE	1300000	340	0.0002615
attB6_attP4	B6	P4	FALSE	1300000	3900	0.0030000
attB6_attP5	B6	P5	FALSE	1600000	92	0.0000575
attB6_attP6	B6	P6	TRUE	1600000	9900	0.0061875

Now we can plot:

#Plot only the att sites that worked well. The full heatmap is shown in supplement.
plot_heatmap <- ggplot(df_heatmap %>% filter(attP!='P4' & attB != 'B4')%>% filter(attP!='P6' & attB != 'B6'), aes(x = attP, y = attB, fill = eff)) + 
  geom_tile(color = 'white', size = 0.25) + 
  geom_tile(data = . %>% filter(match == T), fill = NA, color = 'black', size = 0.5)+
  geom_text(aes(label = scales::label_percent(accuracy = 0.01)(eff)), size = 2.85)+
  scale_fill_viridis_c(trans = 'log',label = scales::percent_format(), breaks = c(0.005, 0.0005,0.00005)) + 
  labs(y = 'attB version\n(∆galK targeting oligo)' , x = 'attP version\n(pInt_chlorR)', fill = 'Efficiency')

plot_heatmap

Fig. 5C - Double Orbit efficiency

Let’s read in the data:

df_eff <- read_csv("../../data/low_throughput_experiments/2022_11_04_double_triple_orbit_att4_data.csv") 

df_eff %>% kable() %>% kable_styling() %>% scroll_box(height = '250px')

cuvette	condition	replicate	LB_count	Kan_count	Chlor_count	Kan_Chlor_count	Carb_count	Kan_Chlor_Amp_count
1	galKattB1_hisAattB1_chlorAttP1_kanAttP1	1	1.9e+07	21000	170000	134	NA	NA
2	galKattB1_hisAattB1_chlorAttP1_kanAttP1	2	1.2e+07	17100	200000	75	NA	NA
3	galKattB2_hisAattB1_chlorAttP2_kanAttP1	1	1.4e+07	6400	37000	61	NA	NA
4	galKattB2_hisAattB1_chlorAttP2_kanAttP1	2	9.4e+06	19000	120000	280	NA	NA
5	galKattB3_hisAattB1_chlorAttP3_kanAttP1	1	9.4e+07	19000	38000	112	NA	NA
6	galKattB3_hisAattB1_chlorAttP3_kanAttP1	2	6.4e+06	18000	45000	90	NA	NA
7	galKattB5_hisAattB1_chlorAttP5_kanAttP1	1	1.3e+07	28000	62000	160	NA	NA
8	galKattB5_hisAattB1_chlorAttP5_kanAttP1	2	1.9e+07	20000	49000	100	NA	NA
9	galK_hisA_metA100_kan_chlor_amp	1	1.3e+07	29000	150000	149	250000	250
10	galK_hisA_metA100_kan_chlor_amp	2	1.8e+07	34000	132000	130	300000	120
11	p265_Kan	1	1.3e+07	16000	NA	NA	NA	NA
12	pInt_only_Kan_Chlor	2	8.2e+06	130	1500	0	NA	NA

Next we can use the single mutant frequencies to calculate the expected double mutant frequencies. We will also get the averages for plotting:

df_eff_meta <- df_eff %>% 
  mutate(kan_eff = Kan_count / LB_count,
         chlor_eff = Chlor_count / LB_count,
         kan_chlor_eff = Kan_Chlor_count / LB_count) %>% 
  mutate(expected_freq =  kan_eff * chlor_eff) %>% 
  group_by(condition) %>% 
  mutate(avg_kan_chlor_eff = mean(kan_chlor_eff))
  
df_eff_meta %>% kable() %>% kable_styling() %>% scroll_box(height = '250px')

cuvette	condition	replicate	LB_count	Kan_count	Chlor_count	Kan_Chlor_count	Carb_count	Kan_Chlor_Amp_count	kan_eff	chlor_eff	kan_chlor_eff	expected_freq	avg_kan_chlor_eff
1	galKattB1_hisAattB1_chlorAttP1_kanAttP1	1	1.9e+07	21000	170000	134	NA	NA	0.0011053	0.0089474	7.10e-06	9.90e-06	6.70e-06
2	galKattB1_hisAattB1_chlorAttP1_kanAttP1	2	1.2e+07	17100	200000	75	NA	NA	0.0014250	0.0166667	6.30e-06	2.38e-05	6.70e-06
3	galKattB2_hisAattB1_chlorAttP2_kanAttP1	1	1.4e+07	6400	37000	61	NA	NA	0.0004571	0.0026429	4.40e-06	1.20e-06	1.71e-05
4	galKattB2_hisAattB1_chlorAttP2_kanAttP1	2	9.4e+06	19000	120000	280	NA	NA	0.0020213	0.0127660	2.98e-05	2.58e-05	1.71e-05
5	galKattB3_hisAattB1_chlorAttP3_kanAttP1	1	9.4e+07	19000	38000	112	NA	NA	0.0002021	0.0004043	1.20e-06	1.00e-07	7.60e-06
6	galKattB3_hisAattB1_chlorAttP3_kanAttP1	2	6.4e+06	18000	45000	90	NA	NA	0.0028125	0.0070313	1.41e-05	1.98e-05	7.60e-06
7	galKattB5_hisAattB1_chlorAttP5_kanAttP1	1	1.3e+07	28000	62000	160	NA	NA	0.0021538	0.0047692	1.23e-05	1.03e-05	8.80e-06
8	galKattB5_hisAattB1_chlorAttP5_kanAttP1	2	1.9e+07	20000	49000	100	NA	NA	0.0010526	0.0025789	5.30e-06	2.70e-06	8.80e-06
9	galK_hisA_metA100_kan_chlor_amp	1	1.3e+07	29000	150000	149	250000	250	0.0022308	0.0115385	1.15e-05	2.57e-05	9.30e-06
10	galK_hisA_metA100_kan_chlor_amp	2	1.8e+07	34000	132000	130	300000	120	0.0018889	0.0073333	7.20e-06	1.39e-05	9.30e-06
11	p265_Kan	1	1.3e+07	16000	NA	NA	NA	NA	0.0012308	NA	NA	NA	NA
12	pInt_only_Kan_Chlor	2	8.2e+06	130	1500	0	NA	NA	0.0000159	0.0001829	0.00e+00	0.00e+00	0.00e+00

Now we can plot everything:

#negative control value
pInt_only_expected <- (df_eff_meta %>% filter(condition == 'pInt_only_Kan_Chlor'))$expected_freq

#filter data to only plot double mutants (single mutant was done as control, triple mutant was also attempted)
plot_eff <- ggplot(df_eff_meta %>% filter(condition != 'galK_hisA_metA100_kan_chlor_amp' & condition != 'p265_Kan' & condition != 'pInt_only_Kan_Chlor'),
       aes(x = condition, y = kan_chlor_eff , group = factor(replicate))) + 
  geom_hline(yintercept = pInt_only_expected, linetype = 2, color = 'light gray')+
  geom_linerange(aes(x = condition, ymin = expected_freq, ymax = kan_chlor_eff), position = position_dodge(width = 0.5), color = 'light gray', linetype = 3)+
   geom_point(aes(y = expected_freq), shape = 4, position = position_dodge(width = 0.5),color = 'light gray')+ #expected frequency as X
   geom_point(shape = 21, position = position_dodge(width = 0.5), fill = 'white',color = 'light gray') + #observed frequency as open circle
  geom_point(data = . %>% filter(replicate == 1), aes(y = avg_kan_chlor_eff), size = 2)+
  scale_y_log10(labels = scales::label_log())+
  scale_x_discrete(labels = c('B1-P1\nB1-P1','B2-P2\nB1-P1','B3-P3\nB1-P1','B5-P5\nB1-P1'))+
  labs(x = '∆galK∆hisA att site composition',y = 'kanR / chlorR efficiency')

plot_eff

Create Fig. 5

theme_set(theme_figure())


fig_5_multi <- plot_grid(plot_heatmap + guides(fill = 'none'), plot_eff, ncol = 1,
                   align = 'hv', axis = 'lr', scale = 1,
                   labels = c('B','C'))


fig_5_multi

save_plot("../../figures/r_pdf_figs/main_figs/fig_5_multi.pdf", fig_5_multi, base_width = 2.5, base_height = 5)

sessionInfo()

## R version 4.2.0 (2022-04-22)
## Platform: x86_64-apple-darwin17.0 (64-bit)
## Running under: macOS Big Sur/Monterey 10.16
## 
## Matrix products: default
## BLAS:   /Library/Frameworks/R.framework/Versions/4.2/Resources/lib/libRblas.0.dylib
## LAPACK: /Library/Frameworks/R.framework/Versions/4.2/Resources/lib/libRlapack.dylib
## 
## locale:
## [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
## 
## attached base packages:
## [1] stats     graphics  grDevices utils     datasets  methods   base     
## 
## other attached packages:
##  [1] kableExtra_1.3.4  cowplot_1.1.1     viridis_0.6.2     viridisLite_0.4.1
##  [5] knitr_1.41        forcats_0.5.2     stringr_1.5.0     dplyr_1.1.0      
##  [9] purrr_0.3.5       readr_2.1.3       tidyr_1.2.1       tibble_3.1.8     
## [13] ggplot2_3.4.0     tidyverse_1.3.2  
## 
## loaded via a namespace (and not attached):
##  [1] httr_1.4.4          sass_0.4.4          bit64_4.0.5        
##  [4] vroom_1.6.0         jsonlite_1.8.3      modelr_0.1.10      
##  [7] bslib_0.4.1         assertthat_0.2.1    highr_0.9          
## [10] googlesheets4_1.0.1 cellranger_1.1.0    yaml_2.3.6         
## [13] pillar_1.8.1        backports_1.4.1     glue_1.6.2         
## [16] digest_0.6.30       rvest_1.0.3         colorspace_2.0-3   
## [19] htmltools_0.5.4     pkgconfig_2.0.3     broom_1.0.1        
## [22] haven_2.5.1         scales_1.2.1        webshot_0.5.4      
## [25] svglite_2.1.0       tzdb_0.3.0          timechange_0.1.1   
## [28] googledrive_2.0.0   generics_0.1.3      farver_2.1.1       
## [31] ellipsis_0.3.2      cachem_1.0.6        withr_2.5.0        
## [34] cli_3.4.1           magrittr_2.0.3      crayon_1.5.2       
## [37] readxl_1.4.1        evaluate_0.18       fs_1.5.2           
## [40] fansi_1.0.3         xml2_1.3.3          textshaping_0.3.6  
## [43] tools_4.2.0         hms_1.1.2           gargle_1.2.1       
## [46] lifecycle_1.0.3     munsell_0.5.0       reprex_2.0.2       
## [49] compiler_4.2.0      jquerylib_0.1.4     systemfonts_1.0.4  
## [52] rlang_1.0.6         grid_4.2.0          rstudioapi_0.14    
## [55] labeling_0.4.2      rmarkdown_2.18      gtable_0.3.1       
## [58] DBI_1.1.3           R6_2.5.1            gridExtra_2.3      
## [61] lubridate_1.9.0     fastmap_1.1.0       bit_4.0.5          
## [64] utf8_1.2.2          ragg_1.2.5          stringi_1.7.8      
## [67] parallel_4.2.0      vctrs_0.5.2         dbplyr_2.2.1       
## [70] tidyselect_1.2.0    xfun_0.35

Figure 5: Multi ORBIT

E. coli ORBIT 2023

Scott H. Saunders

Notes

Fig. 5B - Orthogonal heatmap

Fig. 5C - Double Orbit efficiency

Create Fig. 5