Submit Crab RNA for sequencing

[grace-ac]

Cool beans! How?

Related: this is my notebook post on my selection for pooling... Has explanation of pooling schema shown below.

Pooling schema image (pools are different colors (n=12)):

Note 1: The Day 9 infected (dark orange) and uninfected (light orange) pools were picked such that there are 4 crabs/pool (because the other pools are 4 crabs) and based on the Avg sq day 01 value and I picked the lowest for the uninfected and the highest for the infected (orange boxes corresponding to day 9 pools)

Note 2: I can try and do this in R if that is preferred.

[sr320]

For the 12 libraries indicate the quantity or RNA added to the 12 pools.

On Thu, Jun 7, 2018 at 3:12 PM grace-ac notifications@github.com wrote:

Cool beans! How?

Related: this is my notebook post https://github.com/grace-ac/grace-ac.github.io/blob/master/_posts/2018-06-03-samples-for-seq.md on my selection for pooling... Has explanation of pooling schema shown below.

Pooling schema image (pools are different colors (n=12)): [image: img] https://camo.githubusercontent.com/c7748919496ad3081f694ac54edd9d097c72f0af/687474703a2f2f6f776c2e666973682e77617368696e67746f6e2e6564752f736361706861706f64612f67726163652f437261622d70726f6a6563742f32303138303630372d73616d706c65732d666f722d7365712e706e67

Note 1: The Day 9 infected (dark orange) and uninfected (light orange) pools were picked such that there are 4 crabs/pool (because the other pools are 4 crabs) and based on the Avg sq day 01 value and I picked the lowest for the uninfected and the highest for the infected (orange boxes corresponding to day 9 pools)

Note 2: I can try and do this in R if that is preferred.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/RobertsLab/resources/issues/285#issuecomment-395581967, or mute the thread https://github.com/notifications/unsubscribe-auth/AEPHt54Ar32dD9wLP2HRQfDKDDFrsV_Cks5t6aUbgaJpZM4UfKt_ .

[grace-ac]

Are you talking about Qubit results for the samples? Adding up the amount of RNA in each proposed pool?

[sr320]

Amount of rna of individuals should be equal. How will you do that?

On Thu, Jun 7, 2018 at 3:43 PM grace-ac notifications@github.com wrote:

Are you talking about Qubit results for the samples? Adding up the amount of RNA in each proposed pool?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/RobertsLab/resources/issues/285#issuecomment-395588656, or mute the thread https://github.com/notifications/unsubscribe-auth/AEPHt4SSgLTPnrt13lyTY0WRK9zdhR9wks5t6axkgaJpZM4UfKt_ .

[grace-ac]

Current thoughts: Send out aliquots of each sample such that each will have at least more than 20 ng/µl (#184 ) We request RNA normalized to a minimum of 20ng/uL with a total volume of 50uL.

Pick a goal RNA amount... 30 ng/µl (?)

Example: original sample concentration = 1.71 ng/µl (qubit result) divide 30 ng/ul by 1.71 ng/ul, gives 17.54 So, in 17.54 ul of this particular sample, there will be 30 ng of RNA.

The sample to send off needs a total volume of 50 ul, so I'd just add 32.54 ul of 0.1% DEPC-treated water.

Am I thinking about this right?

[sr320]

Yes, generally correct. Think about it this way.

You will send 12 tubes.

Provide details on how you create these tubes. So someone else could complete.

On Thu, Jun 7, 2018 at 5:28 PM grace-ac notifications@github.com wrote:

Current thoughts: Send out aliquots of each sample such that each will have at least more than 20 ng/µl (#184 https://github.com/RobertsLab/resources/issues/184 ) We request RNA normalized to a minimum of 20ng/uL with a total volume of 50uL.

Pick a goal RNA amount... 30 ng/µl (?)

Example: original sample concentration = 1.71 ng/µl (qubit result) divide 30 ng/ul by 1.71 ng/ul, gives 17.54 So, in 17.54 ul of this particular sample, there will be 30 ng of RNA.

The sample to send off needs a total volume of 50 ul, so I'd just add 32.54 ul of 0.1% DEPC-treated water.

Am I thinking about this right?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/RobertsLab/resources/issues/285#issuecomment-395607363, or mute the thread https://github.com/notifications/unsubscribe-auth/AEPHt8VRIpMTUr0eXr4piCwPR5HzUNhJks5t6cTygaJpZM4UfKt_ .

[grace-ac]

Four tubes (samples) per pool.

Since they'll all be in the same tube eventually, can I combine the four tubes into one, then run Qubit to get RNA amount, then do the math to aliquot out 30ng into a new tube, then add DEPC-treated H20 to get to 50µl... ?

[sr320]

You can, but do not.

On Fri, Jun 8, 2018 at 11:42 AM grace-ac notifications@github.com wrote:

Four tubes (samples) per pool.

Since they'll all be in the same tube eventually, can I combine the four tubes into one, then run Qubit to get RNA amount, then do the math to aliquot out 30ng into a new tube, then add DEPC-treated H20 to get to 50µl... ?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/RobertsLab/resources/issues/285#issuecomment-395852166, or mute the thread https://github.com/notifications/unsubscribe-auth/AEPHtzGBRtNxFeEzg_5CVf4MgKvv2kqEks5t6sVngaJpZM4UfKt_ .

[grace-ac]

Haha, ok.

do the math for each tube so that I am aliquoting out, say, 10 ng of RNA. Add those aliquots into a new tube (total, 40ng of RNA) then add enough DEPC-H20 to get a final volume of 50 ul....

[sr320]

Yes, lets discuss once you have done the math. On Fri, Jun 8, 2018 at 11:55 AM grace-ac notifications@github.com wrote:

Haha, ok.

do the math for each tube so that I am aliquoting out, say, 10 ng of RNA. Add those aliquots into a new tube (total, 40ng of RNA) then add enough DEPC-H20 to get a final volume of 50 ul....

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/RobertsLab/resources/issues/285#issuecomment-395855796, or mute the thread https://github.com/notifications/unsubscribe-auth/AEPHtxZp0jaHhOII8S2QF7YDEj2_h6l2ks5t6shngaJpZM4UfKt_ .

[grace-ac]

Amount of uL to have 10ng of RNA from the tube. Pools will have 40 ng of RNA total in a 50ul sample.

is 40ng too little?? Can easily go more.

[grace-ac]

Pretty sure the above is super off... goal is to have 20ng/ul of RNA in a 50ul sample... so I need at least 1000ng of RNA in a 50ul sample.

Issue #184
From kubu4: Received the following info from the sequencing facility:

We request RNA normalized to a minimum of 20ng/uL with a total volume of 50uL.`

At least 250ng of RNA per tube within a pool (4 samples per pool; 250x4=1000ng). Most samples don't have this much RNA.

[sr320]

@kubu4 it would appear that all these samples need to be precipitated before pooling- could you take a look and go ahead and precipitate if you see fit.

[kubu4]

Ideally, the pooled samples (e.g. a total of three tubes, each with >1000ng of RNA) would be best for precipitation.

However, looking through this issue, I'm not entirely clear on what's been decided in regards to how the pooled samples will be produced.

Each pool needs at least 20ng/uL of RNA (it can be more) and the final volume of each pool should be at least 50uL (it can be more). However, all pools should be the same concentration and volume.

I think @grace-ac needs to determine what volumes of each sample will be used to contribute an equal amount of RNA (ng) to each pool. Once that is determined, we can make a decision regarding pool concentrations and volumes.

To aid with this, this would be my approach:

• Multiply all sample concentrations for a given pool by the minimum sample volume to determine total yield (@grace-ac did this by multiplying all samples by 50uL - however, the samples probably don't have 50uL in them, as some was used for quantification. Use a more conservative volume - say 45uL).

• Find the sample with the lowest total yield.

• Determine volumes needed of all other samples for a given pool to match the sample with the lowest total yield.

• Determine volume and concentration of subsequent pool, and make decision about precipitation.

[sr320]

@grace-ac - can you work with Sam on this?

On Wed, Jun 27, 2018 at 9:01 AM kubu4 notifications@github.com wrote:

Ideally, the pooled samples (e.g. a total of three tubes, each with

1000ng of RNA) would be best for precipitation.

However, looking through this issue, I'm not entirely clear on what's been decided in regards to how the pooled samples will be produced.

Each pool needs at least 20ng/uL of RNA (it can be more) and the final volume of each pool should be at least 50uL (it can be more). However, all pools should be the same concentration and volume.

I think @grace-ac https://github.com/grace-ac needs to determine what volumes of each sample will be used to contribute an equal amount of RNA (ng) to each pool. Once that is determined, we can make a decision regarding pool concentrations and volumes.

To aid with this, this would be my approach:

-

Multiply all sample concentrations for a given pool by the minimum sample volume to determine total yield (@grace-ac https://github.com/grace-ac did this by multiplying all samples by 50uL - however, the samples probably don't have 50uL in them, as some was used for quantification. Use a more conservative volume - say 45uL).

Find the sample with the lowest total yield.

Determine volumes needed of all other samples for a given pool to match the sample with the lowest total yield.

Determine volume and concentration of subsequent pool, and make decision about precipitation.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/RobertsLab/resources/issues/285#issuecomment-400730047, or mute the thread https://github.com/notifications/unsubscribe-auth/AEPHtzM0CgtBvmZ94Bl4qO8nM70fpw2Lks5uA6wMgaJpZM4UfKt_ .

[grace-ac]

• Pink boxes contain the amounts (ul) to take out of sample such that 37.8 ng of RNA (the lowest amount in all of these samples) is removed (ex: 37.8 ng / 7.32 ng/ul = 5.16 ul)
• RNAday9_ng/ul : the Qubit results for the original sample concentration of RNA
• totalRNA_d9_ng : the Qubit results multiplied by 45ul to get the approx. total amount of RNA in samples

[kubu4]

I see three pink boxes above. Does that mean there are only three pools?

Earlier in this thread I see these discussions about pools:

For the 12 libraries indicate the quantity or RNA added to the 12 pools.

Please perform calculations I mentioned above on a per pool basis.

[kubu4]

Also, and I know this is a bit annoying, but could you also please add a column titled "sample_volume(uL)" and fill in the column with 45? It just adds a bit of clarity for the totalRNA_d#_ng columns.

[grace-ac]

still 12 pools! i understand now. I'll fix and post here when done

[grace-ac]

added more columns for clarity! pink boxes around pools

[kubu4]

Great, thanks!

Next:

• Determine volumes and RNA concentration of pools

[kubu4]

Also, you may as well round the numbers to one or two decimal places, since you won't be able to pipette volumes smaller than 0.01uL. This will also make the spreadsheet easier to view. :)

[grace-ac]

Pink boxes are each pool number; total volume ul (sum of d##_ul_pool); RNA concentration ng/ul (lowest RNA ng multiplied by 4 and divided by total volume)

[kubu4]

Create columns called something like:

d#_total_RNA_ng_pool

This should contain the total amount of RNA in each pool. Use this number to calculate the concentration of each pool (i.e. total amount of RNA in a pool divided by the total volume of that pool).

[kubu4]

However, the bigger issue that I see right now is that none of the pools will end up with more than 1000ng of RNA in them. This is the minimum amount of RNA required by the Northwest Genomics Center (i.e. the Nickerson Lab).

We'll likely need to consider sending samples elsewhere with less stringent submission requirements.

[kubu4]

Sorry, I should not have said "none of the pools" - that's incorrect. Some of the pools will hit the 1000ng mark, but not all.

[sr320]

I think we need to try to stick with UW as service. My suggestion is to add individuals to the pool to hit 1000ng. @grace-ac can you come up with a plan to get pools to hit > 1000ng ?

[kubu4]

One other option is making libraries ourselves...

[grace-ac]

I can definitely try that. I’m doing the final steps of moving out and to a new place today, but I’ll be available and will be working on this!

On Thu, Jun 28, 2018 at 7:54 AM kubu4 notifications@github.com wrote:

One other option is making libraries ourselves...

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/RobertsLab/resources/issues/285#issuecomment-401062889, or mute the thread https://github.com/notifications/unsubscribe-auth/AOPhKkQTqDoJFlS6vRg0Z6h9U1SFxNWhks5uBO3ogaJpZM4UfKt_ .

-- Grace Crandall Graduate Student, Roberts Lab School of Aquatic and Fishery Sciences University of Washington, Seattle (206) 221-0978 <javascript:void(0);> | graceac9@uw.edu

[kubu4]

I'll just throw this out there, since I requested the quote yesterday:

• Genewiz pricing is $3840 for 12 library preps, run on a single HiSeq lane (not sure which model).
• Required input RNA is 500ng per library.
• Sequencing yield would be ~20-25 million reads per library.

[grace-ac]

Added column with total pooled RNA (ng). Results highlighted in pink boxes. Got value by multiplying the lowest amount of RNA (ng) in the pool by 4 (4 samples per pool).

[grace-ac]

The only things that I can think about in terms of trying to get 1000ng of RNA...

• evaporate 0.1% DEPC-treated water
• try to pipette some 0.1% DEPC-treated water out

Both seem questionable. Evaporating could degrade RNA and pipetting could accidentally remove the RNA that I'm trying to save...