---
title: "GO Annotations"
author: "Steven Roberts"
date: "`r format(Sys.time(), '%d %B, %Y')`"  
format:
  md:
    toc: true
    toc-depth: 2
editor: 
  markdown: 
    wrap: 72
---




```{r setup, include=FALSE}
knitr::opts_chunk$set(
  echo = TRUE,         # Display code chunks
  eval = TRUE,         # 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
  comment = ""         # Prevents appending '##' to beginning of lines in code output
)
```


Want to start with grabbing protein with specific GOs 

see also https://www.ebi.ac.uk/QuickGO/annotations


# Variables
```{r setup2, include=TRUE}
# Global R options
knitr::opts_chunk$set(echo = TRUE)

# Define key paths and tool directories
 
OUT_DIR <- "../output/27-Apul-pheno-annot/"
evalue <- "1E-20"
fasta <- "../data/Apulchra-genome.pep.faa"

# Export these as environment variables for bash chunks.
Sys.setenv(
  OUT_DIR = OUT_DIR,
  evalue = evalue,
  fasta =fasta
)

```


# Aerobic respiration (GO:0009060)

```{r, engine='bash'}
GO="0009060"

curl -H "Accept: text/plain" "https://rest.uniprot.org/uniprotkb/stream?format=fasta&query=%28%28go%3A"${GO}"%29%29+AND+%28reviewed%3Atrue%29" -o "${OUT_DIR}"SwissProt-GO:"${GO}".fa

head "${OUT_DIR}"SwissProt-GO:"${GO}".fa

echo "Number of Proteins"
grep -c ">" "${OUT_DIR}"SwissProt-GO:"${GO}".fa

/home/shared/ncbi-blast-2.15.0+/bin/makeblastdb \
-in "${OUT_DIR}"SwissProt-GO:"${GO}".fa \
-dbtype prot \
-out "${OUT_DIR}"SwissProt-GO:"${GO}"


/home/shared/ncbi-blast-2.15.0+/bin/blastp \
-query $fasta \
-db "${OUT_DIR}"SwissProt-GO:"${GO}" \
-out "${OUT_DIR}"Apul_blastp-GO:"${GO}"_out.tab \
-evalue "${evalue}" \
-num_threads 42 \
-max_target_seqs 1 \
-max_hsps 1 \
-outfmt 6 \
2> "${OUT_DIR}"blast_warnings"${GO}".txt

head "${OUT_DIR}"Apul_blastp-GO:"${GO}"_out.tab

echo "Number of hits"

wc -l "${OUT_DIR}"Apul_blastp-GO:"${GO}"_out.tab
```


# Oxidative phosphorylation (GO:0006119)


```{r, engine='bash'}
GO="0006119"

curl -H "Accept: text/plain" "https://rest.uniprot.org/uniprotkb/stream?format=fasta&query=%28%28go%3A"${GO}"%29%29+AND+%28reviewed%3Atrue%29" -o "${OUT_DIR}"SwissProt-GO:"${GO}".fa

head "${OUT_DIR}"SwissProt-GO:"${GO}".fa

echo "Number of Proteins"
grep -c ">" "${OUT_DIR}"SwissProt-GO:"${GO}".fa

/home/shared/ncbi-blast-2.15.0+/bin/makeblastdb \
-in "${OUT_DIR}"SwissProt-GO:"${GO}".fa \
-dbtype prot \
-out "${OUT_DIR}"SwissProt-GO:"${GO}"


/home/shared/ncbi-blast-2.15.0+/bin/blastp \
-query $fasta \
-db "${OUT_DIR}"SwissProt-GO:"${GO}" \
-out "${OUT_DIR}"Apul_blastp-GO:"${GO}"_out.tab \
-evalue "${evalue}" \
-num_threads 42 \
-max_target_seqs 1 \
-max_hsps 1 \
-outfmt 6 \
2> "${OUT_DIR}"blast_warnings"${GO}".txt

head "${OUT_DIR}"Apul_blastp-GO:"${GO}"_out.tab

echo "Number of hits"

wc -l "${OUT_DIR}"Apul_blastp-GO:"${GO}"_out.tab
```


# Canonical glycolysis (GO:0061621)

```{r, engine='bash'}
GO="0061621"

curl -H "Accept: text/plain" "https://rest.uniprot.org/uniprotkb/stream?format=fasta&query=%28%28go%3A"${GO}"%29%29+AND+%28reviewed%3Atrue%29" -o "${OUT_DIR}"SwissProt-GO:"${GO}".fa

head "${OUT_DIR}"SwissProt-GO:"${GO}".fa

echo "Number of Proteins"
grep -c ">" "${OUT_DIR}"SwissProt-GO:"${GO}".fa

/home/shared/ncbi-blast-2.15.0+/bin/makeblastdb \
-in "${OUT_DIR}"SwissProt-GO:"${GO}".fa \
-dbtype prot \
-out "${OUT_DIR}"SwissProt-GO:"${GO}"


/home/shared/ncbi-blast-2.15.0+/bin/blastp \
-query $fasta \
-db "${OUT_DIR}"SwissProt-GO:"${GO}" \
-out "${OUT_DIR}"Apul_blastp-GO:"${GO}"_out.tab \
-evalue "${evalue}" \
-num_threads 42 \
-max_target_seqs 1 \
-max_hsps 1 \
-outfmt 6 \
2> "${OUT_DIR}"blast_warnings"${GO}".txt

head "${OUT_DIR}"Apul_blastp-GO:"${GO}"_out.tab

echo "Number of hits"

wc -l "${OUT_DIR}"Apul_blastp-GO:"${GO}"_out.tab
```

# Tricarboxylic Acid Cycle (GO:0006099)

```{r, engine='bash'}
GO="0006099"

curl -H "Accept: text/plain" "https://rest.uniprot.org/uniprotkb/stream?format=fasta&query=%28%28go%3A"${GO}"%29%29+AND+%28reviewed%3Atrue%29" -o "${OUT_DIR}"SwissProt-GO:"${GO}".fa

head "${OUT_DIR}"SwissProt-GO:"${GO}".fa

echo "Number of Proteins"
grep -c ">" "${OUT_DIR}"SwissProt-GO:"${GO}".fa

/home/shared/ncbi-blast-2.15.0+/bin/makeblastdb \
-in "${OUT_DIR}"SwissProt-GO:"${GO}".fa \
-dbtype prot \
-out "${OUT_DIR}"SwissProt-GO:"${GO}"


/home/shared/ncbi-blast-2.15.0+/bin/blastp \
-query $fasta \
-db "${OUT_DIR}"SwissProt-GO:"${GO}" \
-out "${OUT_DIR}"Apul_blastp-GO:"${GO}"_out.tab \
-evalue "${evalue}" \
-num_threads 42 \
-max_target_seqs 1 \
-max_hsps 1 \
-outfmt 6 \
2> "${OUT_DIR}"blast_warnings"${GO}".txt

head "${OUT_DIR}"Apul_blastp-GO:"${GO}"_out.tab

echo "Number of hits"

wc -l "${OUT_DIR}"Apul_blastp-GO:"${GO}"_out.tab
```




# Summary of blast hits

```{r, engine='bash'}
wc -l "${OUT_DIR}"*tab
```