The Answers You’ve Been Looking For

We’ve tested and verified 158 unique antibodies corresponding to 111 human, 62 mouse, and 47 non-human primate markers due to cross-reactivity of some antibodies. The full list can be found in our “In-house Validated Antibodies” downloadable spreadsheet.

Our human custom panel backbone was inspired by Hartmann et al., our mouse custom panel backbone was based on Allen & Hiam et al., and our non-human primate custom panel backbone was inspired by Bjornson-Hooper et al. For each panel, we modified channel placements to allow for easier customization and changed markers to improve immune subset identification. In doing so, we tested and verified each new marker and its channel placement in each panel.

More details on changes made to each panel can be found in our Panel Verification Whitepaper.

We first determined optimal channels (per marker) based on antigen abundance. Next, we determined antibody concentration by testing five different antibody concentrations in a serial dilution. Depending on the nature of the marker, healthy cells were either unstimulated or stimulated with an appropriate biological stimulus, such as PHA or PMA. Finally, spillover was assessed by looking at signals in other channels. To confirm the panel works to detect all major immune cell populations, frequencies of all major immune cell lineages and subsets were measured in PBMCs from 8 healthy subjects.

This is described in finer detail in the Antibody verification section beginning on page 16 of our Panel Verification Whitepaper.

Using a process very similar to how we validated our base panels, we verify the optimal staining concentration at which each metal-conjugated antibody demonstrates a detectable and accurate signal on the mass cytometer, while minimizing background signal and spillover into other channels. Once titrated and validated individually, they are added to the panel and the full panel is then validated.

Yes, we can work with any purified antibody that is stored in a protein-free format (i.e. without BSA or other added proteins). For surface markers, we can work with both IgG and IgM antibodies. For intracellular markers, we recommend using IgG antibodies as they tend to better penetrate into the cell. We do not work with Fab fragments due to incompatibility with our conjugation procedure.

Metal conjugation is performed in-house and the newly metal-conjugated antibody is tested and optimized on in-house PBMC samples. If your marker of interest is not expressed on PBMCs, a cell line or other sample type should be provided or can be purchased for panel validation purposes.

Typically 2 weeks from receipt of all antibody reagents. For some cases, such as those where stimulation is required for marker expression, antibody testing and verification can take up to 4 weeks and this will be communicated in advance.

All antibodies are tested and verified in-house by our team, not outsourced. For new antibodies, pre-conjugated antibodies are used when commercially available. If conjugation is required, it is performed in-house by our team, not outsourced, to ensure high performance and fast turnaround. All new antibodies, whether pre-conjugated or conjugated in-house, are tested and verified by our team.

Currently, we do not support the addition of peptide tetramers into our panels.

Yes, conditionally, and no.

Yes – Transcription factors and other highly expressed intracellular markers can easily be detected. Our base panels typically include transcription factors such as Foxp3, T-bet, and TCF1 and intracellular functional proteins like Granzyme B and KI-67. We have also validated a number of other transcription factors like HIF1a and RORgt, intracellular proteins like Perforin and X, and even metabolic markers like mTOR and GAPDH.

Conditionally – Some cytokines, such as IFNg and TNFa, can be detected in activated cells without additional stimulation. However, most cytokines typically require cell stimulation in the presence of a secretion blocker (like Brefeldin A) in order to build up sufficient protein for detection. Our scientists can help you determine whether prior stimulation may be necessary depending on your cytokine of interest.

No – Chemokines are often produced in low quantities not well-suited for cytometry based detection. While chemokine production cannot be easily detected, chemokine receptor expression is detectable and can provide information about which immune subsets are responsive to chemokine gradients.

We identify the appropriate antibody concentration for markers on our panel using a combination of stain index, arcsinh ratios, and spillover consideration. The verification method that we use depends on the type of marker–phenotypic, those used to identify cell types, or functional, those used to characterize cell state. For phenotypic markers, we determine the optimal concentration using the stain index, calculating the separation between positive and negative populations to find the optimal concentration that minimizes background noise and ensures an accurate signal. In Figure A above, the stain index was calculated for Tbet at varying concentrations, and 3 µg/ml was selected as the concentration with the best separation between positive and negative populations, and with the background noise minimized.

For functional markers that don’t have a strong separation between positive and negative populations, we use the arcsinh ratio to compare antibody staining in for a marker before and after stimulation. To calculate the arcsinh ratio, we arcsinh-transform the median intensity values (to make them easier to compare) and look at the difference between these transformed values for stimulated and unstimulated conditions. In Figure B above, the unstimulated (on the left) and stimulated (on the right) staining conditions are shown. The arcsinh ratio was then calculated for selected concentrations and 3 µg/ml was selected as the optimal concentration.

Additionally, if the separation of the positive signal of a custom marker is not clear and may have spillover from another channel, we also assess for spillover using a “mass minus one” (MMO) control, where we omit one marker at a time to see if it appears in a channel where it shouldn’t. We then choose the antibody concentration that provides a strong signal in the intended channel with minimal background staining in other channels.

Through these methods, we establish the optimal concentration for each marker on the panel and subsequently validate it on control samples, comparing our results with reported flow cytometry data to ensure consistency and reliability.

Yes, with one specific condition: the customer must provide approval of or changes to the gating scheme within 24 hours of receipt of the Sample & Gating QC report, before Teiko analyzes the samples. If the customer approves the gating scheme and later wishes to modify the gating scheme after analysis, it would require a re-analysis, incurring an additional cost.

The answer depends on the confidence you need, and the variability you can accept. Variability is expressed by the coefficient of variation (CV) is simply the standard deviation divided by the mean. The higher this number, the more “variable” the measurement. The lower the number, the less “variable” the measurement. Intuitively, a population that appears a lot, say 10% of the time, needs fewer cells than a population that occurs 0.001% of the time.

Luckily, this question has been addressed in the work of Keeney et al. To wit, “for cell-based assays such as flow cytometry, a simple calculation can be used to determine the size of the database/sample that will provide a given precision: r = (100/CV)2; where r is the number of events meeting the required criterion, and CV is the coefficient of variation of a known positive control.”

For a population that’s at the 0.1% level, you need 10,000,000 events to detect with a 1% CV, and 400,000 events to detect with a 5% CV. And using Keeney’s formula, you need 10,000 events of that specific population at the 1% level, and 400 events at the 5% level.

Intermediate Monocytes (inMono) cells are 0.47% of non-granulocytes in Teiko’s internal data. We would need 188 inMono cells to get past a 5% CV, and we actually collected 302 cells in our internal data. As a result, we ended up with a 4.63% intra-run CV. In case you are wondering, all populations measured using Teiko’s standard panel have met the accepted CV criteria of 25-30%.

For more details and to check out the full calculations, read our article.

Our internal R&D data shows that with Whole Blood you get the same quality information about non-granulocytes as with PBMCs, plus additional insights into granulocyte populations (neutrophils, eosinophils, basophils). Teiko scientist Dr. Jolien Sweere reviews the data in our webinar: Whole blood or PBMCs for immune profiling?

For prospective studies, we recommend collecting whole blood into standard vacutainers containing, in order of preference, heparin, citrate, or EDTA and using our simple two-step, 20 minute TokuKit for on-site processing. This process is faster, easier, and cheaper than PBMC isolation, has no centrifugation steps, and yields high-quality cytometry data.

For retrospective studies, we accept viably cryopreserved PBMCs or whole blood samples that have undergone red blood cell lysis and fixation using PROT1, or Stable-Lyse, Stable-Store.

We cannot accept whole blood collected using an RNA or DNA isolation tube, nor serum samples as these lack intact cells required for cytometry-based analyses.

At Teiko Bio, we’ve worked with a wide range of sample types across human, mouse, and non-human primates. This includes PBMCs, whole blood, bone marrow aspirate, TILs, cell therapy product, and even dissociated fresh tumor and tissue. For a full list of sample types, download our Sample Guidelines.

Not for our spectral flow cytometry assays, but we do fix PBMCs before staining for our mass cytometry assays and see no impact on the dynamic range of most functional markers. For our mass cytometry assays, we fix PBMCs with paraformaldehyde (PFA) before antibody staining because our panels are optimized on such samples. Hartmann et al (Cell Reports, 2019) tested how PFA-fixation before (PFA-stained) or after antibody staining (live-stained) affects mass cytometry immune profiling results. They found that immune lineage frequencies are very similar between live-stained and PFA-stained samples, with an r=0.94 correlation. Hartmann also saw that PFA-fixation does not change the dynamic range of most functional markers; however, a small subset of antigens (e.g. CCR7, CD11b) showed decreased staining. The authors concluded that PFA-fixation before antibody staining has minimal effects on immune profiling data.

Our validation data show the minimum input for accurate detection of all major immune cell populations is 50K viable cells. Minimum tumor and tissue sample size depends on immune infiltration, however a sample volume of 350 mm3 should yield a sufficient number of immune cells for analysis [Allen et al., 2020]. While these are the minimum volumes for detection, we recommend 2mL of whole blood or 1 million viable PBMCs for most samples.

No, sample collection on CyTOF is a destructive process, meaning the cells cannot be recovered after acquisition.

Yes, please let us know when planning your project how much of each sample you would like us to return and where to ship the remaining samples. Thawed PBMC samples can be separated into aliquots and viably cryopreserved for return shipment.

Yes, if samples arrive at our facility within 48 hours of collection. Blood stored at room temperature for 24 to 48 hours shows modest changes in marker expression, but dramatic changes by 72 hours. Temperature fluctuations during shipping and delivery delays are other factors that can impact fresh blood samples. For this reason, we highly recommend fixing blood on-site using our TokuKit and shipping samples on dry ice to preserve marker expression and sample quality during shipping.

At the moment we only include short-term storage of samples (less than 6 months) as part of an analysis project with Teiko. Long-term storage of samples (greater than 6 months) can be added for a nominal fee to a larger analysis plan for ongoing and prospective clinical trials. Cryopreserved samples are stored in liquid nitrogen and fixed samples are stored at -80°C.

Gating your samples is one of the most crucial parts of data analysis. The example below shows how without a good gating strategy you would identify up to 30% false-positive events to be CD8+, when in reality this signal is contamination from sticky cells such as platelets or erythrocytes. Gating might be straightforward for smaller, 3-8 color panels, but becomes highly complex for 20+ marker panels identifying over 30 immune populations.

Our gating starts by excluding debris, dead cells, platelets and red blood cells. In a next step, we remove Basophils, Eosinophils and other Granulocytes. This allows us to report the remaining immune cell frequencies as a percentage of non-Granulocytes, making it easy for you to compare percentage numbers between samples.

Check out the flow plots below for an example of how our gating strategy helps identify clean populations of true CD4+ and CD8+ T-cells within CD3+ non-Granulocytes.

We tailor, or adjust, gates between patients because each individual has a unique immune setpoint, meaning their immune cells may express slightly higher or lower levels of various protein markers at baseline. This causes the populations to shift on the plots as you cycle between patients and the gates must be moved to accommodate this shift on a per-patient basis. Once the gates are set for each patient, we do not shift them in between time points or conditions. This ensures we are able to capture shifts in expression of state markers across time and conditions based on each patient’s unique baseline.

Manual gating is the process of identifying immune cell populations based on the protein markers they are known to express, based on existing literature. For example, we know that T cells express CD3 and B cells express CD19 so we can plot the cells using these two markers as our axes to identify these two populations. (See Fig A)

We can then drill down into each population to identify subsets. For example, we can go into the T cell gate and look at another set of two markers like CD4 and CD8. We continue to do this going down into CD4 or CD8 populations until we’ve identified all of the different T cell subsets. (See Fig B)

Then we can go back to our B cells in the CD19+ gate and use other markers to drill down to B cell subsets. Then into the third gate and use other markers to identify populations like NK cells, Dendritic Cells, and Monocytes by looking at expression of other protein markers unique to those cell types (e.g. CD56 for NK cells, CD14 and CD16 for monocytes, etc.). This requires extensive knowledge of marker expression across all cell types to ensure you’re correctly identifying each population.

We and others (e.g. Rahman, Tordesillas and Berin, Cytometry Part A 2016) have seen that eosinophils are difficult to identify with mass cytometry, even when markers specific to eosinophils (such as SIGLEC 8) are included in the panel. During our validation studies, we noted eosinophils tend to bind nonspecifically to some antibodies for which they don’t express the antigen, which can lead to eosinophils being confused for other cell types (e.g. platelets, neutrophils, etc). We also observed that eosinophil population frequencies are more sensitive to storage conditions (e.g. measured eosinophil frequencies drop sharply if blood is stored for 5 hours at room temperature before fixation), whereas other populations are more stable. Therefore, Teiko decided that unless specifically requested by a customer, we use SIGLEC8 as a ‘dump gate’ to remove eosinophils and increase confidence in other granulocyte populations, rather than report on eosinophils themselves.

While heparin treatment before antibody staining can improve eosinophil detection, that same treatment negatively affects performance of several key state marker antibodies (i.e. Ki67, TCF1, FOXP3 and Tbet). Our team of experts can discuss options if you are particularly interested in eosinophil frequencies for your study.

Note that in any case, we cannot report on eosinophil state markers, as their intracellular granules interfere with antibody binding.

We completed rigorous tests as part of our CLIA registration process. We analyzed peripheral blood mononuclear cells (PBMCs) from 9 healthy donors, and looked and four measures. For each measure, we set a coefficient of variation (CV) threshold, based on the literature. As a refresher, the coefficient of variation is simply the standard deviation divided by the mean. In general, a smaller CV is better.

• Inter-run: Same sample, different runs
  • 3.52% versus CV threshold of 25% criteria
  • one PBMC sample was divided into three replicates and running on CyTOF on the same day
• Intra-run: Same sample, same run
  • 6.66% versus 30% CV criteria
  • three replicates from one PBMC sample were run on different days
• Inter-operator: Same sample, different operators
  • 4.83%, 7.36% versus 25% CV criteria
  • two replicates processed by different operators were compared
• Stability: Same sample, across time
  • -0.01% versus +/- 25% change between days
  • four PBMC samples were split, stored, and analyzed on days 1 and 22

Want to read more?

Our full validation report also includes a description of all protocols and methods, assay design including step-by-step gating strategy, and more.

5 hours: collected blood can be left out at room temperature in the vacutainer for up to 5 hours before processing with the TokuKit. We found that 100% of immune cell subsets are preserved when TokuKits are processed within 5 hours, but that delay to 7 hours results in loss of several important immune lineages. Please note that we do recommend processing within 2 hours post-collection when possible. You can read more about these findings here.

The TokuKit is Teiko’s whole blood collection and preservation kit that customers can use for their research studies and clinical trials. Recognizing how involved isolation of peripheral blood mononuclear cells (PBMCs) can be, Teiko developed the TokuKit to make sample preservation for immune profiling faster and easier. Processing blood with the TokuKit takes at most 20 minutes and 3 steps, and preserves granulocytes for extra immune lineage information! For more information, visit https://teiko.bio/tokukit/.

The TokuKit contains a blood collection vacutainer, and one or two bottles with cell preservation reagents. You collect the blood in the vacutainer, transfer 2 mL of the blood to the cell preservation reagent bottle, and if needed, quench the fixation reaction with the second preservation reagent. Then, the blood can be stored at -80C or shipped to Teiko on dry ice. It is as simple as 1-2-3!

Teiko currently offers 4 variants of the TokuKit. They differ from each other in the cell preservative used (Stable-Lyse, Stable-Store, or PROT1) and the anticoagulant in the blood collection vacutainer (EDTA or sodium heparin). The Teiko team can help you with selecting the TokuKit version that works best for your research needs – contact us through the button below!

Blood samples that were processed with the TokuKit need to be stored at -80C long-term or on dry ice short-term. Sample shipment to Teiko requires dry ice. Samples can be kept in the original TokuKit container.

When we start working on a study with you, we check to see if we can use reagents with the same lot throughout the project. If that is not possible (for example because the project goes on for longer than the buffer expiration date), we will do a ‘bridging experiment’ when we switch lots to ensure the new lot will not impact immune profiling results. Within a lot, we have tested inter-kit variance and saw very low variance between kits for all tested immune cell lineages. Learn more about our inter-kit variability data by clicking the button below.

The expiration date for the TokuKits is dependent on the manufacturing date of the Stable-Lyse buffer or the PROT1 buffer in the kit. Typically, TokuKits have a minimum shelf life of six months when shipped.

To our lab address: 675 Arapeen Drive Suite 301, Salt Lake City, UT 84108. ATTN: Li-Chun Cheng or Teiko Lab. TokuKits can be received Tuesday – Friday 9 am – 5 pm.

Teiko currently does not offer TokuKits for non-human samples, although they will likely work well on non-human primate samples.

Teiko currently does not offer TokuKits (or other collection kits) for tissues other than blood.

Reach out to us at info@teiko.bio! We’ll send you ordering instructions if you are an existing customer, or if you have never worked with us before, we’ll set up an introductory conversation to learn more about your needs.

Teiko has tested stability at -80C up to 12 months, and up to 24 months using accelerated reagent stability, in fixed whole blood samples from two healthy donors. We observed a high concordance in frequencies of immune cell populations (R=1), when compared to samples processed 1 week after freezing.

Maybe, but we have not tested the use of TokuKits for conventional (non-spectral) flow cytometry internally yet. However, two papers published in peer-reviewed journals have shown that it is possible to fix blood using the Stable-Lyse/Stable-Store (Nguyen et al.) and PROT1 (Silva et al.) buffers that we use in our TokuKits and run the samples on conventional flow. These papers found that some markers are not impacted by fixation, while others are, meaning that panels need to be created carefully.

To explore the feasibility of using these buffers for spectral flow cytometry, Teiko is currently testing our 25-marker spectral flow cytometry panel on whole blood samples fixed with our Stable-Lyse/Stable-Store TokuKits. The results will be posted on our website in Q2:24.

No, although we have not tested this in-house. We have not found any publications that have successfully fixed whole blood samples using the buffers in our TokuKits and used those samples for scRNA-seq. Therefore, we do not recommend using whole blood fixed with our TokuKits for scRNA-seq.

No, you don’t need to send all 2 mL of collected blood. The minimum volume we require is 0.5 mL of whole blood. That volume still allows us to detect all populations other than the most rare. In healthy donors, like those used in our CLIA validation report, the only populations we could not reliably detect in 2 mL of blood are transitional dendritic cells, mast cells, plasmablasts, and CD4 TEMRA cells. That said, analyzing 0.5 mL of blood would likely also exclude CD56hi NK cells, double-positive T cells, and plasmacytoid dendritic cells. Note that these populations could be detectable in non-healthy donors with altered immune composition of higher total immune cell counts per mL of blood.

Our kits are optimized for fixing 2 mL of blood, which can be aliquoted after fixation. If you want to fix less than the requested 2 mL of blood, please contact us to discuss custom solutions.

Read more about which populations we look at, their frequencies, and which cells would pass our event threshold at 2mL, 1mL, and 0.5 mL inputs of whole blood.

We conducted an inter-kit precision study and found a total average coefficient of variation (CV) of 5.76% in immune cell population frequencies for all 33 populations analyzed. All populations fell under our 25% CV cutoff. To determine inter-kit precision, we used samples from 3 donors, split one sample from each donor into 3 replicates, and fixed each of the 9 samples with a different Stable-Lyse Stable-Store Heparin TokuKit. We then analyzed the samples using our CLIA-validated Standard whole blood mass cytometry panel.

State markers are specific proteins that indicate the functional state or condition of a cell at a point in time. Some examples of different states of cells include activation, exhaustion, and proliferation. Some state markers on our standard Peripheral Blood Mononuclear Cell (PBMC) panel are Ki67, a protein all cells produce when replicating, and PD-1, a checkpoint protein expressed on exhausted T cells.

For each cell analyzed, we measure the level of expression of all 25+ proteins on our panel, including all state markers. Data for state markers included in our panels and our web app are reported as frequency and a median channel value (MCV). The frequency represents the proportion of cells within each cell subset expressing that state marker above a set threshold. We say these cells are “positive” for this state marker. This can answer the question “What proportion of T cells are exhausted?” with “37% of T cells are positive for PD-1 and therefore are exhausted.”

The MCV represents the intensity of the signal detected for a specific metal isotope associated with a particular antibody tag on a cell. It is similar to median fluorescence intensity, or MFI, in flow cytometry. This can answer the question “Does the amount of PD-1 expressed on T cells increase or decrease with treatment?” with “The amount of PD-1 expressed on T cells decreases with treatment because the MCV value for PD-1 on T cells drops from 5.6 to 4.8 after treatment.” This type of marker expression level information is important since higher PD-1 protein expression correlates with a more exhausted phenotype (PNAS 2013).

We evaluate state marker expression as an MCV for a given marker in the (1) marker-positive cell population and (2) parent population. This allows you to quantify the level of expression within cells that are known to express the state marker of interest and across the entire parent population, including cells with no or low expression of the state marker.

In our WebApp, ‘top gate’ refers to the cell population used as the denominator to calculate the relative frequency of an individual immune cell subset. For granulocytes, the top gate is usually total live cells; for non-granulocytes, the top gate is usually total non-granulocytes. For example: if a patient has 45% neutrophils as percent of top gate, it means that 45% of that patient’s live cells are neutrophils. If another patient has 12% CD4 central memory cells, then 12% of that patient’s non-granulocytes are CD4 central memory cells. The reason we use different top level gates is that if all populations were measured as % of live cells, there would be populations so small it would be difficult to plot them or see clear differences between samples. For example, plasmablasts can be ~0.22% of non-granulocytes, but ~0.09% of total live cells. There are two places where you can check what an individual population’s top gate is: a) Teiko clarifies it in the Scope of Work or Work Order (the project contract); b) you can find it in the About Project > Overview > Subset Hierarchy image.

At Teiko, we used unsupervised clustering to identify distinct Treg functional states associated with better survival in 40 patients with urothelial carcinoma.

We initially applied both gated and unsupervised clustering analysis to identify immune features associated with response. Gated analysis found that higher frequencies of regulatory T cells (Treg) were associated with better survival, a surprising result given that Treg usually suppress tumor-killing immune cells. Unsupervised clustering split the Treg population within the dataset into two distinct clusters. Only one of those was associated with response, and that cluster showed Treg were inactivated. More inactivated Tregs potentially results in stronger anti-tumor activity and greater response to immunotherapy for patients. This crucial Treg population would have been really difficult, if not impossible, to find with manual gating alone. Click the button below to see how we identified this unique Treg cluster!

Yes, you get access to .FCS and Gating-ML files for your project. FCS is the standard file format for flow cytometry experiments. With GatingML, this means you can review the gated data for your project.