# Producing DNA sequencing procedure with MinION

## End goal
- Extracting DNA from a sample with a procedure that is easy and cheap for non-experts to follow.
- Preparing the DNA for use in an ONT minION.

## Overview
- First step was to find a suitable sample candidate. We decided to use material which we assumed had a high consentration of DNA.
- After collecting a suitable sample we took it to the lab and followed different procedures, described later, to extract DNA.
- After acquiring DNA, we sent it to SINTEF's lab at NTNU for preparation of sequence library.
- When the prepared DNA was sent back, we proceded with using the minION to sequence it.

In total we prepared 8 samples all with slightly different procedures. The details of each experiment is noted in the excel file "dna-prodecure.xlsx". Of the 8 samples, we used two for minION sequensing.

## 1. Batch, 2 samples
On the first batch we collected sediment from a nearby beach. We went through the [general procedure](#General-procedure).

After sending to SINTEF, we got this result back:

| Prøve | ng/µL | A260/230 | A260/280 |
|----------|----------|-------------|-------------|
| 1 | 4,055 | 0,77 | 1,5 |
| 2 | 2,016 | 0,92 | 1,62 |

After
| Prøve | ng/µL | A260/230 | A260/280 |
|----------|----------|-------------|-------------|
| 1 | 1,522 | 0,49 | 1,05 |
| 2 | 2,808 | 0,52 | 1,44 |

These had too low concentration of DNA to be able to use for the minION. SINTEF suggested we proceed with pipetting the solution up and down instead of shaking the tube by hand. They also said that the amount of AMPure was more than enough, but we assumed this would not have any impact on the quality of the remaining DNA because of the washing procedure later on. They also recommended us to extend the bead beating time and reduce the amount of liquid in the initial step. For the next batch of samples we decided to send a sample to SINTEF taken from after bead beating and centrifuging such that we could decide whether the cell lysis worked or not.
We would also proceed with Elution Buffer (EB) for the next batch of samples when performing the elution step in which we only used distilled water.

## 2. Batch, 3 samples

To acheive better result we decided to try using an specific elution buffer (Elution Buffer from Qiagen) instead of water and 75% ethanol instead of 95%. We collected more beach sediment, compost soil and crab stomach. We hoped that compost and crab would contain higher concentration of DNA. The result of this is shown here:
| | Nanodrop | | | Qubit | Flongle | miniON |
|:------------------------------:|:-----------:|:-----------:|-------------|------------|----------------|--------------|
| Prøve | ng/µL | A260/230 | A260/280 | ng/µL | 500 ng (µL) | 1 µg (µL) |
| 2. Sedimentprøve | 11,588 | 0,45 | 1,76 | Too low | | |
| 3. Sedimentprøve (rent DNA) | 9,83 | 0,62 | 1,14 | too low | | |
| 4. Krabbepave | 1743,205 | 0,65 | 1,09 | 47,2 | 10,6 | 21,2 |
| 5. Krabbepave (rent DNA) | 79,041 | 0,23 | 1,05 | 0,9 | | |
| 6. Kompostjord (rent DNA) | 481,374 | 0,52 | 1,3 | 7,32 | | |
| 7. Kompostjord (rent DNA) | 1145,958 | 0,59 | 1,29 | 16,3 | 30,7 | 61,3 |

Only sample 4 and 7 gave good enough result to be used. They were prepared by sintef and returned with these concentrations:
| Prøve | ng/µL | A260/230 | A260/280 |
|----------|----------|-------------|-------------|
| 1 | 1,522 | 0,49 | 1,05 |
| 2 | 2,808 | 0,52 | 1,44 |

Using this we follow the procedure as described in Genomic DNA by Ligation (SQK-LSK109)-flongle (1).pdf at the bottom at page 4. The notes we took are in the same folder.

For **sample 4** the minion produced almost no reads of high enough quality for further use. The report is in the minION folder as [NanoPlot-Report-jordrprove.html](miniON/NanoPlot-Report-jordprove.html). For **sample 7** tried to reduce the quality threshold, which resulted in more reads but of bad quality.

## 3. Batch, 1 sample
| | Nanodrop | | | Qubit |
|----------------------------------------------|-------------|-------------|-------------|------------|
| Prøve | ng/µL | A260/230 | A260/280 | ng/µL |
| 1. Ren AMPure prøve 1 | 1366,014 | 0,49 | 1,13 | 6,88 |
| 2. Fortynnet AMPure prøve 2 | 32,272 | 0,43 | 1,39 | too low |
| 3. Magnet 2 min prøve 1 | 4051,403 | 0,93 | 1,2 | 8,2 |
| 4. Fortynnet AMPure, magnet 2 min prøve 2 | 4411,256 | 0,8 | 1,23 | 7,8 |
| 5. Overflødig DNA prøve 1 | - | - | - | 5,2 |

For the last batch, we took another crab stomach and made two main samples (prøve 1 and prøve 2 in the table above). For both samples we followed [modified procedure](#Modified procedure) which was provided by SINTEF. As described in the specific procedure in the minION folder, we added pure AMPure to the first sample, but for the second sample we diluted it to 1:4 (40 µl AMPure + 160 µl distilled water).
The results in the table above show that most of the DNA remains in the liquid instead of close to the magnet. Diluting the AMPure didn't show to be helpful either, so the conclusion was that we could go back to the original method.

## General procedure
Collect sample -> Bead beating ~1min -> Centrifuge: ~1 min 2000G -> Extraction of liquid part -> Add AMPure -> Shake -> Magnet -> Exctract liquid -> Washing with ethanol x2 -> Add elution buffer -> Magnet -> Extract liquid -> DNA

## Modified procedure
1. Tilsetter X µL AMPure beads til DNAet, pipetterer opp og ned 10 ganger. Trenger ikke å være kraftig.
2. Lar rørene stå i 5 minutter uten risting.
3. Setter rørene på magneten i 2 minutter før jeg fjerner væsken.
4. Imens rørene er på magneten tilsetter jeg etanol og lar dette stå på magneten i 30 sekunder.
5. Fjerner så etanol.
6. Repeterer vasketrinnet igjen, fortsatt med rørene på magneten.
7. Fjerner så etanolen, jeg også med en mindre pipette for å sørge for at all etanol er fjernet.
8. Tar så rørene av magneten og tilsetter EB buffer. Her pipetterer jeg 10 ganger til beadsene er godt blandet med EB-bufferen, ikke kraftig.
9. Lar så rørene stå i 2 minutter, uten risting.
10. Setter rørene på magneten igjen og pipetterer væsken med DNA til et nytt rør.

## Sequencing with minION
After preparing a DNA sample with sequencing library it was time to perform the sequencing with the MinION. We used the MinKNOW software to run the sequencing and get the analysis. [How to load a single flow cell](https://www.youtube.com/watch?v=_SyN3fZ3G9Q) shows how the flow cell was loaded. SINTEF recommended we add 14 µl elution buffer to 7 µl of the prepared DNA sample, then take 1 µl of this dilution and further dilute with 4 µl elution buffer. This would then replace the 5 µl of DNA sample described in the protocol. This method was used for the first sequencing procedure (crab stomach), but for the second one (compost soil) we directly used 5 µl of the pure DNA sample.
For the first sequencing (crab stomach), we used the default setting in minKNOW which set the quality score threshold at 9, meaning that reads that didn't meet that level of quality were rejected. This left us with very few reads such that the fastq file at the end was too small to pass the pipeline of analysis. We ran this sequencing for 24 hours and set the minimum read length at 1000 bases per read.
The second sequencing (compost soil) ran for 72 hours, the minimum read length was set to 10(?) and the quality score adjusted to 6. The result was a significantly bigger fastq file, but another issue was that the length of most of the reads was way too small for the file to pass through the pipeline.

## Conclusion
Our method of preparing a DNA sample showed to work to some extent, as the cell lysis worked and we had a relatively high concentration of DNA in the remaining liquid after bead beating and centrifuging. It seemed as if the AMPure didn't bind itself to the DNA strands, which is to be worked further on.