GeCKO library PCR for sequencing: how to maintain diversity

[Huiyi Chen]

Hi all,

I have read some of the older threads here about how to do the PCR to generate a library for sequencing. I couldn't find an exact protocol to follow, although the guidelines are scattered in various messages.

1. I would like to check the distribution of my plasmid library. I've seen that others use 10ng to set up PCR1. Do you then have to expand this single PCR to 7 tubes for PCR2 to maintain diversity?

2. For a 10kb plasmid, 10ng is 10^9 molecules. However, for gDNA, 10ug is 10^6 molecules. Can we use the same number of amplifications cycles for gDNA and plasmid PCRs?

Thanks,

Huiyi

[Julia Joung]

Hi Huiyi,

We have recently put together a screening Nat Protocols paper that might be helpful for your project. The details for how to set up PCR for plasmid library amplification and subsequent analysis for both the plasmid and gDNA amplification steps are included in the protocol.

Best,

Julia

--
You received this message because you are subscribed to the Google Groups "Genome Engineering using CRISPR/Cas Systems" group.
To unsubscribe from this group and stop receiving emails from it, send an email to crispr+unsubscribe@googlegroups.com.
To post to this group, send email to cri...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Huiyi Chen]

Hi Julia,

Thanks for the paper. I've gone through the protocol, and it's very different from the original protocol in the GeCKO library paper. Unfortunately, I have already started on the library construction earlier this year, and my reagents are not compatible with the newest protocol. I'll just have to be conservative with regards to the PCR (better to have too many PCRs than too few).

Would you have any comments on why the lab has gone with a new polymerase and a slightly different set of primers?

Thanks,

Huiyi

On Saturday, May 13, 2017 at 12:03:13 AM UTC+8, Julia Joung wrote:

Hi Huiyi,

We have recently put together a screening Nat Protocols paper that might be helpful for your project. The details for how to set up PCR for plasmid library amplification and subsequent analysis for both the plasmid and gDNA amplification steps are included in the protocol.

Best,

Julia

On Fri, May 12, 2017 at 3:32 AM, Huiyi Chen <hui...@gmail.com> wrote:

Hi all,

I have read some of the older threads here about how to do the PCR to generate a library for sequencing. I couldn't find an exact protocol to follow, although the guidelines are scattered in various messages.

1. I would like to check the distribution of my plasmid library. I've seen that others use 10ng to set up PCR1. Do you then have to expand this single PCR to 7 tubes for PCR2 to maintain diversity?

2. For a 10kb plasmid, 10ng is 10^9 molecules. However, for gDNA, 10ug is 10^6 molecules. Can we use the same number of amplifications cycles for gDNA and plasmid PCRs?

Thanks,

Huiyi

--
You received this message because you are subscribed to the Google Groups "Genome Engineering using CRISPR/Cas Systems" group.

To unsubscribe from this group and stop receiving emails from it, send an email to crispr+un...@googlegroups.com.

[Julia Joung]

Hi Huiyi,

We chose to switch to NEBNext for library NGS because NEBNext has higher fidelity and can amplify more diverse sequences, making it more desirable for NGS. The primers provided allow for 1-step PCR instead of 2-step PCR. 1-step PCR reduces the number of steps required, reagent cost, and potential bias.

Best,

Julia

To unsubscribe from this group and stop receiving emails from it, send an email to crispr+unsubscribe@googlegroups.com.

[Huiyi Chen]

Hi Julia,

Can you share information on how much better the NGS results are with the NEBNext compared to the Herculase II polymerase? I'm guessing you get more useful results? 

I can't do anything about the current library, but it would be useful for future work.

Thanks,

Huiyi

[Julia Joung]

Hi Huiyi,

We did not do a direct comparison in the lab, but NEB has done their own comparison that is available on their NEBNext product page. We have used the PCR protocol outlined in the Nat Protocol paper for two of our recent screens, and it has worked well.

Best,

Julia

[Gunjan Kumar]

Hi Julia,

I had a somewhat related question regarding the nature protocols paper. For lentivirus generation, your group chose to use lipofectamine 2000 to transfect the 293FT cells instead of lipo 3000. Was there any particular reason for doing so since lipofectamine 3000 seems to be cheaper and is a more efficient reagent? I am just wondering if I could use lipo 3000 without affecting the library. 

Thanks!

Gunjan

[Julia Joung]

Hi Gunjan,

Good point. We chose lipofectamine 2000 for lentivirus production in the protocol paper because our lab has been using lipofectamine 2000 to produce lentivirus for the published screens. I don't see any reason why using lipo 3000 would affect the library, and it is certainly worth trying because the lipo 3000 protocol suggests that lipo 3000 will produce a much higher titer.

Per your suggestion, I'm planning on testing lipofectamine 3000 for lentivirus production this week, and will keep you updated.

Best,

Julia

To unsubscribe from this group and stop receiving emails from it, send an email to crispr+unsubscribe@googlegroups.com.

[Gunjan Kumar]

Thanks Julia! Please do let me know how it goes. Look forward to hearing back from you. 

[Julia Joung]

Hi Gunjan,

We just did a side-by-side comparison of lipofectamine 2000 and 3000 (see attached). We adjusted the L3000 protocol to fit the Nat Protocols, using 15uL of P3000 enhancer and 17.5uL of L3000 for a T25 transfection with the same amount of plasmids. We replaced 2.5mL of media with 1250uL of L3000 mix, changed the entire volume of media after 5h, and harvest virus at 48h post transfection. In conditions with caffeine, we added caffeine to 2mM ~18-24h after transfection. We measured the functional titer by transducing HEK cells with different amounts of virus supernatant (as detailed in the Nat Protocols).

As you can see from our results, L3000 does indeed produce higher lentiviral titer than L2000, and because we use less, it is also cheaper per transfection (~40% of the cost of L2000). Our lab is in the process of switching over to L3000 for our lentiviral prep.

Thanks again for the suggestion!

Best,

Julia

To unsubscribe from this group and stop receiving emails from it, send an email to crispr+unsubscribe@googlegroups.com.

[Aaron]

Hi Julia,

This looks great !!!

Can you be so kind to provide me with the detail protocol of Trasfecting a T25 flask for lentiviral production ?

best,

Aaron 

[Julia Joung]

Hi Aaron,

I'm attaching the protocol for using lipofectamine 3000 to transfect a T25 flask. Please let me know if you have any more questions!

Best,

Julia

To unsubscribe from this group and stop receiving emails from it, send an email to crispr+unsubscribe@googlegroups.com.

[jian...@gmail.com]

Is anyone compared the differences of using 293T and 293FT for virus preparation? If this has been discussed before, please post the related link again.

Thanks

[Julia Joung]

Hi,

Based on my experience, 293FT cells produce much higher lentivirus titer.

Best,

Julia

--

[Haiyin Liu]

Hi Julia,

thanks for your input regarding the new GeCKO protocols. I have a question regarding gDNA input for the 1-step PCR.

In the old protocol, one only needed 100µg gDNA per sample for the PCR1, and from that could use a small amount of PCR1 product to set up multiple PCR2s for desired technical replicates with different index combinations. However, with the new 1-step PCR, do I need N times 100µg gDNA? I.e. for 3x technical replicates I would need to set up 3x 1-step PCRs all requiring 100gDNA each?

Best regards,

Haiyin

[Julia Joung]

Hi Haiyin,

For harvesting the screen, we recommend harvesting >500 cells/sgRNA in the library (e.g. >50 million cells for a 100,000 sgRNA library) and using all of the gDNA from the harvest for the PCR. We usually do not do technical replicates for the PCR and instead rely on screening infection replicates for different bioreps of the library, which would be barcoded separately (see our screening Nat Protocol for more details).

Best,

Julia

--

[Gunjan Kumar]

Hi Julia,

Glad this worked! Thanks for letting me know the results from your trials :)

Cheers,

Gunjan

[Julia Joung]

No problem - thanks again for the suggestion!

To unsubscribe from this group and stop receiving emails from it, send an email to crispr+unsubscribe@googlegroups.com.

[Gunjan Kumar]

You're welcome! I did have a couple of other question as well:

1. I'm actually using John Doench's Brunello library for my screen. But I imagine since it's still in the lentiCRISPRv2 system, the lentiviral production and screening protocol would be the same as that of GeCKOv2?

2. I have also noticed that the viral concentration step has been omitted from the new protocol (I had initially been using the GeCKOv1 library and protocol). Is it no longer necessary with the v2 vector system? 

Thanks!

Gunjan

[Haiyin Liu]

Hi there,

Julia: Thanks for your reply! That makes sense and helps a lot.

Gunjan: The v2 system is designed to have an increased titre over v1. While my labmate used v2 and had to use the viral suepernatant neat, I'm using the GeCKOv2 system and actually have to dilute the SN to get the desired MOI. From what I've heard from collaborators it's common to dilute v2 SN 1:5-1:15 for transduction.

Cheers,

Haiyin

[Julia Joung]

Hi Gunjan,

To answer your questions:

1. Yes because the Brunello library is still in the lentiCRISPRv2 system (like GeCKOv2), our screening protocol also applies to the Brunello library. I would actually recommend using the Brunello library over the GeCKOv2 library because it uses a more updated on/off-target scoring algorithm.

2. Haiyin is correct in saying that since the GeCKOv2 system typically produces higher titer than v1, for most cell lines viral concentration is no longer necessary.

Best,

Julia

To unsubscribe from this group and stop receiving emails from it, send an email to crispr+unsubscribe@googlegroups.com.

[sara ahrabi]

Hi Julia,

Many thanks for explaining the library prep steps for NGS. I am also in the library prep step following harvesting DNA from my two samples.

I did some optimisation using different concentrations of DNA and got the best result using 0.5 micrograms DNA per PCR. 

This is the best I could get and I have my products at around 260-270 bp but always this smaller non-specific band is present. So I decided to go ahead with these conditions and then gel extract my bands.

I have now completed the PCRs using the 10 forward and 2 reverse primers as indicated in your 2017 Nature protocol paper (https://www.nature.com/articles/nprot.2017.016).

I ended up doing 180 PCRs for one of my samples and 120 PCRs for the other one. 

My question is how would be best to gel purify 300x50 microlitres of PCR product?

Thanks

Sara

To unsubscribe from this group and stop receiving emails from it, send an email to cri...@googlegroups.com.

[Julia Joung]

Hi Sara,

In the protocol paper, we also describe how to best purify large amounts of PCR product using Zymo columns. Let me know if you have any questions about that part of the protocol.

Best,

Julia

To view this discussion on the web visit https://groups.google.com/d/msgid/crispr/c68d3c75-3f02-48b7-bd0f-18956bed2e88%40googlegroups.com.

[Minoo karimi]

Hi Julia, 

I am following your Nature protocol to make a custom sgRNA library. If I am not mistaken, you prepared your library with 500 fold coverage, right? did you make any glycerol stock for your library? If Yes, Could you please share your protocol with me? Thanks a lot!

Minoo

On Friday, May 12, 2017 at 12:03:13 PM UTC-4 juliaj...@gmail.com wrote:

Hi Huiyi,

We have recently put together a screening Nat Protocols paper that might be helpful for your project. The details for how to set up PCR for plasmid library amplification and subsequent analysis for both the plasmid and gDNA amplification steps are included in the protocol.

Best,

Julia

On Fri, May 12, 2017 at 3:32 AM, Huiyi Chen <hui...@gmail.com> wrote:

Hi all,

I have read some of the older threads here about how to do the PCR to generate a library for sequencing. I couldn't find an exact protocol to follow, although the guidelines are scattered in various messages.

1. I would like to check the distribution of my plasmid library. I've seen that others use 10ng to set up PCR1. Do you then have to expand this single PCR to 7 tubes for PCR2 to maintain diversity?

2. For a 10kb plasmid, 10ng is 10^9 molecules. However, for gDNA, 10ug is 10^6 molecules. Can we use the same number of amplifications cycles for gDNA and plasmid PCRs?

Thanks,

Huiyi

--
You received this message because you are subscribed to the Google Groups "Genome Engineering using CRISPR/Cas Systems" group.

To unsubscribe from this group and stop receiving emails from it, send an email to crispr+un...@googlegroups.com.