FAQ - STOmics

Q Why is there a flipping problem using automatic registration?

A

● A flip problem can occur when the microscope photo picture of the tissue and the gene expression matrix are aligned, and the flip situation is related to the chip sequencing scheme and the image output of the microscope.

● SS chips start sequencing from the lower left corner during sequencing, the coordinate origin (0,0) of sequencing is in the lower left corner, and the coordinate origin of the imaging are in the upper left corner, so the image and expression matrix will be reversed by default in the automatic image process.

● However, after imaging, the output image may be in the same orientation as the image seen by the lens, or the output image may be mirrored with the orientation of the tissue seen by the lens due to the different configuration of the microscope's software.

● How to judge in advance whether the image output by your microscope needs to be flipped before QC/analysis?

○ The Stereo-seq technology uses a chip to capture, and the chip is opaque, so no matter whether the microscope is placed upright or inverted, the microscope needs the lens to take pictures of the tissue staining image. Therefore, when acquiring microscope images, as long as it is confirmed that the direction of the lens facing the tissue is consistent with the direction of the image output by the microscope software, and there is no mirror image, the problem of inversion and matching will not occur during registration.

● If there is a mirror image of the output image of the microscope and the orientation of the tissue on the chip, you can try to adjust the microscope configuration, or use PhotoShop, ImageJ and other image processing software to flip the image, and then QC.

翻转问题-配图

Q What does the overall image processing procedure of Stereo-seq analysis look like?

A

stereo-seq_image_processing_overview

Q Format and requirements of the original input FASTQ file to SAW.

A

STOmics SAW analysis process supports Q40 FASTQ and Q4 FASTQ files as input parameters.

Q40: A quality system that describes the quality of all sequenced bases with 41 quality values. Q40 FASTQ is the common form of PE FASTQ, with data appearing in pairs, usually named <lane>*_1.fq.gz and <lane>*_2.fq.gz. In general, the Q40 FASTQ file usually has a "read" field, such as lane_read_1.fq.gz and lane_read_2.fq.gz. In a few cases, there may be no "read" in the file name, then you can check whether it is in pairs, or check the file format.

## Q40 read 1
@V350156489L4C001R00100001484/1
TCTGCAGCCAACATGGACAGATCCTTTTAGAACTT
+
D>DC<CBEADEABCB(AADCDD2DD"DBE*$F'C'

## Q40 read 2
@V350156489L4C001R00100001484/2
CTATGAAACACACTATCCTCAATCGGCTCCTTAATTTCAATACCAGCCGT
+
ECFCCB?CAECEBDBCCFDFEFDF?<FCF>B?2EC=C?C<CE(AA=;B5B

Q4: A quality system that describes the quality of all sequenced bases with 4 quality values. Q4 FASTQ comes as a set of 16 or 64 individual files, usually named <Lane>_<barcode>_<splitIndex>.fq.gz. In a few cases, there may be no barcode in the file prefix, you may distinguish the barcode from the directory name as below. Although it's not accurate, Q4 FASTQ is previously often called SE FASTQ. In fact, it is still paired-end sequencing, but when FASTQ is written out of the machine after sequencing, it is output as a set of single files, which looks the same as SE data. This format encodes the information in read1 and writes it into the readID of read2. At the same time, the sequence quality information is recorded in the form of Q4 to reduce storage.

requirements

## Q4 read
@FP300000513L1C002R00400000218 CE242DF29A57 97D26
GTGTAGTGAACCCCATGGTAGTTTTCTGATTGTTGTTAAAAAAAATGACTTAACATATTACATGGACACTCAATAAAAATGTTTTATTTCCTGTTGAAAA
+
FFFFFFFFFFFF8F8FFFFFFFFFFFFF8FFFFFFFFF8FF8FFF8FFFFFFF,FFFFFFFFFFF8FFFFFF8F8F,F8FFFFFF,FFFFFFFFFF,FFF

Q Does the max length used for annotating alternative splicing sequences need to be consistent to the read length?

A

Does parameter --sjdOverhang used for STAR builds of reference genome indexes need to be consistent with the sequencing read length? If there are variable sequencing read lengths, e.g. 50bp, 100bp, 150bp, will SAW choose different STAR indices automatically during analysis?

At present, there's no distinct index built for different read lengths. Reference genomes used for submitted standard analysis tasks on the STOmic Cloud platform are uniformly built with the STAR default value for a read length of 100, that is --sjdOverhang 99.
Using the different lengths in genome indexing can specify the range of adjacent genome sequences around the annotated splicing sites, which is the maximum length used for identification of alternative transcript splicing.
The impact of using different lengths is limited to very few splicing sites that are susceptible to variation of several bases in the alignment of the regions around boundaries of exons/introns. The difference is marginal that there is no need to have a distinct index for each read length.

Reference material:

Q What is the purpose of each DNA clean-up step using beads in the Stereo-seq Transcriptomics Set protocol? What do the beads remove or harvest?

A

Purification Steps	Purposes
0.8X beads (after cDNA release from the tissue)	The high ratio of beads allows binding of cDNA molecules as many as possible, while leaving impurities from the tissue samplein the supernatant.
0.6X beads (after cDNA amplification by PCR)	The beads bind cDNA molecules. Primers and other small DNA fragments will remain in the supernatant and get discarded.
0.55X beads and 0.15 μL beads (after cDNA fragmentation & amplification)	Double selection which removes both larger and smaller fragmented DNA and harvests the intermediate fragments.

Q Can the primary antibody staining solution be incubated with the chip at 4℃ for overnight instead of at room temperature for 45 min?

A

We suggest following the protocol with all antibody incubation steps. Since the antibody needs to be incubated with the Stereo-seq Chip T, please avoid leaving the chip with the antibody at 4℃ overnight.

Q Have other secondary antibodies been tested besides Alexa Fluro Plus 2nd antibodies ?

A

We have only tested Alexa Fluro Plus 2^nd antibodies.

Q Are there any recommended blocking solutions for non-human species?

A

We recommend to use serum and FCR blocking reagent for preparing the blocking solution. For non-human/mouse species, we recommend to substitute FcR blocking reagent with nuclease-free water.

Q Is 3 antibodies with DAPI the maximum co-staining targets we have tested in R&D?

A

Yes, that's the maximum co-staining targets we have tested in R&D.

Q Is Stereo-seq mIF workflow applicable on any chip sizes other than 1cm x 1cm?

A

Currently not, but we are in the process of testing Stereo-seq mIF workflow on other chip sizes besides 1cm x 1cm.