-
Notifications
You must be signed in to change notification settings - Fork 0
/
NCBI_SRA_overtime.Rmd
151 lines (129 loc) · 5.09 KB
/
NCBI_SRA_overtime.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
---
title: "NCBI - SRA over time"
author: "André Rodrigues Soares"
date: "`r Sys.Date()`"
output: html_document
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```
initial code from https://gist.github.com/BenLangmead/545634578b6a80e5bba0e31843923e7a
```{r}
library(tidyverse)
```
```{r}
system('curl https://trace.ncbi.nlm.nih.gov/Traces/sra/sra_stat.cgi > stats.csv')
sra = read_csv("stats.csv",
col_types = cols(
date = col_datetime(format = "%m/%d/%Y"),
bases = col_double(),
open_access_bases = col_double(),
bytes = col_double(),
open_access_bytes = col_double())) %>%
select(date, bases, bytes) %>%
gather("data", "values", -date) %>%
mutate(date = as.Date(as.character(date)),
log_vals = log(values),
days = as.numeric(difftime(as.Date(date),
min(as.Date(date)),units = "days")))
sra_bases = sra %>%
filter(data == "bases")
sra_bytes = sra %>%
filter(data == "bytes")
```
## Trends over time
Linear model fitted as from 2017
```{r}
ggplot(sra,
aes(date, log_vals,
colour = data, fill = data)) +
geom_smooth(method = 'lm',
se = T) +
theme_bw()
m_bases <- lm(log_vals ~ days, data = sra_bases %>%
filter(days > 2000))
summary(m_bases)
model_bases <- nls(values ~ a * exp(b * days),
data = sra_bases,
start = list(a = exp(summary(m_bases)$coefficients[1,1]), b = 0.001))
model_bases
summary(model_bases)
m_bytes <- lm(log_vals ~ days, data = sra_bytes %>%
filter(days > 2000))
summary(m_bytes)
model_bytes <- nls(values ~ a * exp(b * days),
data = sra_bytes,
start = list(a = exp(summary(m_bytes)$coefficients[1,1]), b = 0.001))
model_bytes
summary(model_bytes)
# Predict future values
# Calculate the difference in days between max date and 2030-01-01
days_difference <- as.numeric(difftime(as.Date("2030-01-01"),
max(sra_bases$date), units = "days"))
# Generate future_time
future_time <- seq(max(sra_bases$days),
max(sra_bases$days) + days_difference,
by = 1)
future_values_bases <- predict(model_bases,
newdata = data.frame(days = future_time), # Corrected variable name
interval = "prediction")
future_values_bytes <- predict(model_bytes,
newdata = data.frame(days = future_time), # Corrected variable name
interval = "prediction")
future_df = tibble(days = future_time,
bases = future_values_bases,
bytes = future_values_bytes) %>%
mutate(date = min(sra_bases$date) + future_time) %>%
pivot_longer(names_to = "data", values_to = "values", -c(days, date)) %>%
add_column(type = "prediction")
all_data = bind_rows(sra %>%
add_column(type = "data"),
future_df)
ggplot(all_data,
aes(date, values*0.000000000001,
colour = data,
linetype = type)) +
geom_line() +
geom_vline(xintercept = as.Date("2024-01-01"),
linetype = 'dashed', colour = 'grey',
alpha = .7) +
scale_x_date(date_breaks = "2 years",
date_labels = "%Y") +
scale_y_continuous(labels = scales::comma_format(),
expand = c(0,0)) +
scale_colour_manual(values = c("#CC6677", "#117733", "#CC6677", "#117733"),
labels = c("SRA bases", "SRA bytes", "Predicted SRA bases", "Predicted SRA bytes"),
name = "SRA data & predictions") +
guides(linetype = "none") +
theme_bw() +
theme(axis.title = element_blank(),
axis.text.x = element_text(hjust = 1, vjust = 1, angle = 20),
legend.position = "bottom") +
ggtitle("TBytes and Tbases in the NCBI-SRA since 2007",
subtitle = paste0("Exponential model fitted from data from 2012-11-25 onwards\nLast data point: ", max(sra_bases$date)))
```
```{r}
ggplot(all_data,
aes(date, values*0.000000000001,
colour = data,
linetype = type)) +
geom_line() +
geom_vline(xintercept = as.Date("2024-01-01"),
linetype = 'dashed', colour = 'grey',
alpha = .7) +
scale_x_date(date_breaks = "2 years",
date_labels = "%Y") +
scale_y_log10(labels = scales::comma_format(),
expand = c(0,0),
breaks = c(1, 100, 10000, 500000)) +
scale_colour_manual(values = c("#CC6677", "#117733", "#CC6677", "#117733"),
labels = c("SRA bases", "SRA bytes", "Predicted SRA bases", "Predicted SRA bytes"),
name = "SRA data & predictions") +
guides(linetype = "none") +
theme_bw() +
theme(axis.title = element_blank(),
axis.text.x = element_text(hjust = 1, vjust = 1, angle = 20),
legend.position = "bottom") +
ggtitle("TBytes and Tbases in the NCBI-SRA since 2007",
subtitle = paste0("Exponential model fitted from data from 2012-11-25 onwards\nLast data point: ", max(sra_bases$date)))
```