15 August 2014

DGE, Illumina library prep, haplotype imputation, BEAGLE, and KAPA

Next post Previous post

11 -- 15 August notes

2014-08-11

ApGxL

Ovation DGE for eight samples.

Contacted Ashesh at Nugen Tech Support

uses oligo(T) to initiate the first strand synthesis.  The length of the product is determined by the processivity of the reverse transcriptase.  This is the reason that you get cDNA of varying lengths.  When you fragment the cDNA you will have fragments that are adjacent to the poly(A) and fragment further upstream depending on the size of the initially generated cDNA.   There is no specific selection of just fragments proximal to the poly(A) tail. 
 
I would recommend using 1ug of cDNA as input into the Encore Rapid library system.  You should fragment at least 1.5 ug of cDNA and then purify the fragmented cDNA.  The purified, fragmented cDNA should be quantified and 1ug of the fragmented cDNA used as input into the library construction system.

So - this DGE protocol is not only a single tag per transcript, but simply a reduced length cDNA transcript?!? Not quite as efficient as I thought/hoped.

MtGIRAFFE

Desktop can’t process whole chr8 file - workstation still in shop. Working with smaller test file.

Export to VCF

bcftools view -f PASS -r chr5:1000-6000 -O v -o test.vcf chr5-filtered-set-2014Apr15.bcf

Convert to BEAGLE format using Beagle utility program

cat test.vcf | java -jar vcf2beagle.jar ? BeagleTest

Run BEAGLE

java -Xmx1000m -jar beagle.jar unphased=BeagleTest.bgl.gz missing=? out=BeagleTest

2014-08-12

ApGxL

Post-SPIA protocol for yesterday’s samples
Purified using QIAquick into 32 ul TE

Qubit cDNA assay results

Sample	ng/ul
07-B-CT	69
11-C-CT	104
15-A-CT	77
20-C-CT	64.6
26-C-CT	27.3
07-A-HS	32.7
07-B-HS	0.7
11-C-HS	7.1

Samples 1 - 4 had > 60 ng/ul with 2 ul in 198 working solution, diluted 2x by adding another 200 ul Qubit DNA buffer to get readings.

Ovation library prep

Pooling 16 samples
First step, fragment DNA using Covaris.
For each sample, added volume to get 1500 ng DNA or all DNA available

Number	Sample	stock (ng/ul)	volume for 1.5ug (ul)	actually added (ul)
1	08-D-CT	>60	25	30
2	15-B-HS	26.9	55.8	25
3	05-B-HS	18.0	83.3	25
4	07-B-CT	69	21.74	21.8
5	11-C-CT	104	14.4	14.4
6	15-A-CT	77	19.5	19.5
7	20-C-CT	64.6	23.2	23.2
8	26-C-CT	27.3	55.0	25
9	07-A-HS	32.7	45.9	28
10	11-C-HS	7.13	210.4	28
11	J3-A	104	15	15
12	J3-B	104	15	1
13	J2-A	215	7	5
14	J2-B	215	7	1
15	J1	194	7.7	5
16	J4	34.2	43.8	25

Used a range of input concentrations to assess effect of DNA input mass on library prep
Transferred sample to Covaris tubes and filled to 50ul with Tris
Covaris S2 fragmentation settings: Time 120 seconds, Intensity 5, Duty Cycle 10% and Cycles per burst 200.
Sent aliquot of each sample to Bioanalyzer

2014-08-13

ApGxL

Library prep
concentrated fragmented cDNA using Qiagen QIAquick to get small volume, ~11 ul per sample
8 ul are used on library prep, with target of 1ug of DNA

Discussed questions about DGE from Ashesh’s email with SCH. Possibility of using un-fragmented DNA. In this case, only ~50% of reads would be useful - those that map to 3’ end of transcript. If used paired-end sequencing, might be able to use pair to anchor read to transcript. Might be a problem with using fragments of varying length in Illumina…

Extracted RNA from 12 more samples

UVM-AGTC: Problems with Bioanalyzer for fragmented DNA - possible contaminant that is masking ladder and making it difficult to get clear trace.

2014-08-14

ApGxL

Library prep
DNA concentration after fragmentation and purification

Number	Sample s	tock (ng/ul)
1	08-D-CT	102
2	15-B-HS	132
3	05-B-HS	106
4	07-B-CT	93
5	11-C-CT	94
6	15-A-CT	90
7	20-C-CT	98
8	26-C-CT	27.8
9	07-A-HS	139
10	11-C-HS	0
11	J3-A	41
12	J3-B	1.3
13	J2-A	23
14	J2-B	1.2
15	J1	26
16	J4	63

Low-input samples (11-C-HS, J3-B, J2-B) have about zero DNA - drop from sequencing library.
For input of 500 ng in 7 ul, need concentration of 71 ng/ul. 5 samples (26-C-CT, J3-A, J2-A, J1, J4) are below this level but non-zero. Proceed with them anyway to see what happens with low-input library prep.
As I have ‘extra’ barcodes, decided to pool all the J1 and J2 fragments from last week as they were about the same size into a single sample and include in library prep
concentrated fragmented cDNA using protocol in Encore Rapid Protocol. eluted into 20 ul TE
quantified DNA
- J1-P: 28.9 ng/ul
- J2-P: 27.9 ng/ul
Followed libary prep protocol.

DGE library JSG001

Number	Sample	Barcode	Sequence
1	08-D-CT	L2DR01	AAGGGA
2	15-B-HS	L2DR02	CCTTCA
3	05-B-HS	L2DR03	GGACCC
4	07-B-CT	L2DR04	TTCAGC
5	11-C-CT	L2DR05	AAGACG
6	15-A-CT	L2DR06	CCTCGG
7	20-C-CT	L2DR07	GGATGT
8	26-C-CT	L2DR08	TTCGCT
9	07-A-HS	L2DR09	ACACGA
10	J3-A	L2DR10	CACACA
11	J2-A	L2DR11	GTGTTA
12	J1	L2DR12	TGTGAA
13	J4	L2DR13	ACAAAC
14	J1-P	L2DR14	CACCTC
15	J2-P	L2DR15	GTGGCC

Kappa Library Quant qPCR - called tech support to ask if I could use the kit we ordered for LightCycler 480 with new ABS StepOnePlus (S1P) - Yes! S1P uses a reference dye (Rox High) while the LightCycler did not. Simply need to turn off reference dye in S1P settings. To do this, under Plate Settings select the Assign Target and Samples tab. In bottom-right, set Reference Dye to None - Could result in more jagged curves, not as smooth as with reference dye - For future, order KK4835 - Very good tech support from Kappa!

UVM-AGCT

Meeting with Tim Hunter and Scott Tighe

Discussed sequencing plans
cost of full RNAseq vs DGE
use low EDTA TE Buffer: Tris 10 mM EDTA 0.1 mM
ST specified SPIA product, not cDNA
Qubit and Kappa often consistent. Cheaper and faster to pool samples based on Qubit values, than run single Kappa qPCR. For first run, do Kappa qPCR for all samples then use values to normalize for library prep
should I be using more ants at the extraction step?
Bioanalyzer results wonky; standards not showing up
DNA too concentrated? dilute to under 5 ng/ul
trace pattern still looks okay, consistent with fragmented SPIA product

R

UseR was this week - information via Twitter:

2014-08-15

ApGxL

KAPA Library Quantification qPCR

Diluted library 1:1000 ul in 10mM Tris
- made a 1:2000 dilution (100 ul dilution 1 in 100ul Tris)
Mixed samples

Reagent	Volume (ul)
KAPA master mix	12
H₂O	4
DNA/standard	4

Set-up plate with:
- 6 Illumina standards in triplicate
- Triplicate samples of each dilution for each sample
Ran qPCR
- standard curve
- SYBR green
- 2 hr run
- Passive reference: none
- Cycling: 95C - 5 min, 35x {95C - 30sec, 60C - 45sec}

Results

Set standards to 20, 2, 0.2, 0.02, 0.002 and 0.0002 pM
Recorded “Quantity* from Unknown samples
Could only fit 13 samples on plate

Calculated concentration of undiluted library stock (pM):

\[ C = \overline{quantity} \times \frac{452}{200} \times Dilution^-1 \]

where 200 is the average fragment length.

Number	Sample	pM
1	08-D-CT	198
2	15-B-HS	4,144
3	05-B-HS	257
4	07-B-CT	905
5	11-C-CT	948
6	15-A-CT	1,858
7	20-C-CT	1,840
8	26-C-CT	828
9	07-A-HS	1,154
10	J3-A	674
11	J2-A	554
12	J1	481
13	J4	685

Great! Have amplifiable inserts for all samples!