---
<<<<<<< HEAD
title: "FISH546 Week 1: Blast Assignment"
author: "Sam Cryan"
date: "`r format(Sys.time(), '%d %B, %Y')`" 
output: 
  html_document:
    theme: readable
    toc: true
    toc_float: true
    number_sections: true
    code_folding: show
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
knitr::opts_chunk$set(
=======
title: "FISH546 Assignment 1: BLAST testing"
author: "Sam Cryan"
date: "2023-04-11"
output: html_document
---

This document generates a BLAST database 

```{r setup, include=FALSE}
library(knitr)
opts_chunk$set(
>>>>>>> 15610ea2cc2f623fab2331bcbee37bd6a3a57195
  echo = TRUE,         # Display code chunks
  eval = FALSE,         # Evaluate code chunks
  warning = FALSE,     # Hide warnings
  message = FALSE,     # Hide messages
  fig.width = 6,       # Set plot width in inches
  fig.height = 4,      # Set plot height in inches
  fig.align = "center" # Align plots to the center
)
```

## Setting Up Blast Scripts

<<<<<<< HEAD
=======
# ONLY RUN ON JUPYTER
>>>>>>> 15610ea2cc2f623fab2331bcbee37bd6a3a57195
```{bash}
cd ~/applications
curl -O https://ftp.ncbi.nlm.nih.gov/blast/executables/blast+/LATEST/ncbi-blast-2.13.0+-x64-linux.tar.gz
```
```{bash}
cd ~/applications
 tar -xf ncbi-blast-2.13.0+-x64-linux.tar.gz
```

# Check that blast is installed
```{bash}
~/applications/ncbi-blast-2.13.0+/bin/blastx -h
```



```{bash}
cd /scryan_coursework/scryan-coursework/assignments/data
curl -O https://ftp.uniprot.org/pub/databases/uniprot/current_release/knowledgebase/complete/uniprot_sprot.fasta.gz
mv uniprot_sprot.fasta.gz uniprot_sprot_r2023_01.fasta.gz
gunzip -k uniprot_sprot_r2023_01.fasta.gz
```

```{bash}
~/applications/ncbi-blast-2.13.0+/bin/makeblastdb \
-in ./data/uniprot_sprot_r2023_01.fasta \
-dbtype prot \
-out ./blastdb/uniprot_sprot_r2023_01
```

## Download unknown BLAST sequences
```{bash}
curl https://eagle.fish.washington.edu/cnidarian/Ab_4denovo_CLC6_a.fa \
-k \
> ./data/Ab_4denovo_CLC6_a.fa
```

```{bash}
~/applications/ncbi-blast-2.13.0+/bin/blastx \
-query ./data/Ab_4denovo_CLC6_a.fa \
-db ./blastdb/uniprot_sprot_r2023_01 \
-out ./output/Ab_4-uniprot_blastx.tab \
-evalue 1E-20 \
-num_threads 20 \
-max_target_seqs 1 \
-outfmt 6
```

```{bash}
curl -o uniprot_table_r2023_01.tab  -H "Accept: text/plain; format=tsv" "https://rest.uniprot.org/uniprotkb/stream?compressed=true&fields=accession%2Creviewed%2Cid%2Cprotein_name%2Cgene_names%2Corganism_name%2Clength%2Ccc_induction%2Cgo%2Cgo_p%2Cgo_id%2Cxref_unipathway&format=tsv&query=%28%2A%29%20AND%20%28reviewed%3Atrue%29&sort=gene%20asc"
```


```{bash}
tr '|' '\t' < ./output/Ab_4-uniprot_blastx.tab \
> ./output/Ab_4-uniprot_blastx_sep.tab
```

```{bash}
head -2 ./data/uniprot_table_r2023_01.tab
# wc -l ./data/uniprot_table_r2023_01.tab
```
``` {r load libaries}
library(tidyverse)
library("kableExtra")
```
``` {r read data}
bltabl <- read.csv("./output/Ab_4-uniprot_blastx_sep.tab", sep = '\t', header = FALSE)

spgo <- read.csv("https://gannet.fish.washington.edu/seashell/snaps/uniprot_table_r2023_01.tab", sep = '\t', header = TRUE)
```


``` {r test join}
kbl(
  left_join(bltabl, spgo,  by = c("V3" = "Entry")) %>%
  select(V1, V3, V13, Protein.names, Organism, Gene.Ontology..biological.process., Gene.Ontology.IDs) %>% mutate(V1 = str_replace_all(V1, 
            pattern = "solid0078_20110412_FRAG_BC_WHITE_WHITE_F3_QV_SE_trimmed", replacement = "Ab"))
) %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive"))
```