João Costa

[João Costa @jonnymates]

@AliceFromQueens Let's all admire the uniqueness of hitler, you sad fuck

[Luciano Martelotto 🛠🧬💻🇦🇺 @LGMartelotto]

What if single-cell analysis of 1,000 samples wasn’t just imaginable—but affordable?

Imagine analyzing 1,000 samples at the single-cell level, each containing ~25,000 cells or nuclei. That’s ~25 million cells in one experiment!. Until now, the costs of such an experiment were staggering, making it seem like a dream only billion-dollar projects could afford. But the game has changed and this is no longer science fiction.

Let’s break it down.

For this, let’s choose the RNA Flex family of kits from 10x Genomics. In my unsolicited opinion the best single cell RNA kits out there yet.

To estimate the total cost for single-cell RNA sequencing (scRNA-seq) of 1,000 samples, each containing 25,000 cells or nuclei, we'll consider both library preparation and sequencing expenses, and calculate the cost per cell. For this case, let’s consider we have a large cohort (n=1000) of post vaccine PBMC samples.

Library Preparation Costs:

The average cost of an RNA Flex reaction at core facilities in the USA is $4,492 for four samples, using a four-barcode pooling strategy. This amounts to $1,123 per sample. Since each reaction accommodates ~10,000 cells and the goal is ~25,000 cells per sample, 2.5 reactions are required per sample:

•Cost per sample: $1,123 × 2.5 = $2,807.50
•Total library preparation cost for 1,000 samples: $2,807.50 × 1,000 = $2,807,500

Sequencing Costs:

The ballpark read-pairs/cell most researchers I know are using is at least c 20,000 read-pairs/cell so for ~25,000 cells per sample, we'll need:

•Total reads per sample: ~25,000 cells × 20,000 reads/cell = ~500 million reads
•Total reads for 1,000 samples: ~500 million reads × 1,000 samples = ~500 billion reads

Assuming a sequencing cost of $1.50 per million reads on the Illumina NovaSeq X platform:

•Sequencing cost per sample: ~500 million reads × $1.50 per million reads = $750

•Total sequencing cost for 1,000 samples: ~500 billion reads × $1.50 per million reads = $750,000

Total Project Cost:

•Library preparation: $2,807,500
•Sequencing: $750,000
•Total cost per sample: $2,807.50 (library preparation) + $750 (sequencing) = $3,557.50
•Total cost for 1,000 samples: $2,807,500 + $750,000 = $3,557,500
•Total cost per cell: $3,557,500 / 25,000,00 cells = $0.1423 per cell

Yeah…scary!

Ok…don’t panic, we got you. Recently, @DrJasPlummer @hoheyn and yours truly made an orthogonal approach to perform single cell analysis without sequencing: STAMP: Single-Cell Transcriptomics Analysis and Multimodal Profiling through Imaging (doi: https://t.co/m4OHy3BP5N). Here, we repurposed various imaging platforms like Xenium and CosMx to perform Single Cell Imaging…no sequencing!

What do you think the costs of profiling 25 million cell would be?

Let’s see…

To estimate the total cost for STAMP of 1,000 samples, each containing 25,000 cells or nuclei. For this, let’s consider that we have 50 sub-STAMPs/slide, that’s 100 per run on a Xenium slide and even more in a standard superfrost slide for CosMx, but let’s stick to 50/slide. For the calculations, the costs per run for both Imaging Platforms costs per run are similar.

Total Imaging Platform Runs Needed:

•Number of samples: 1000
•Sub-STAMPs per slide: 50
•Slides per run: 2
•Total slides required = 20 slides.
•Total Imaging Platform runs = 10 runs

Total Cost:

•Cost per Imaging Platform run (2 slides) = $7,500 (average across various cores in USA for 5k and 6k panels)
•Cost per sample = $75
•Total cost for 1,000 samples = $75,000
•Cost per cell: $0.003/cell (0.3 cents per cell)

Now that you’ve seen this, do me a favor—step away for a bit. Take a walk, ride your bike, or do whatever inspires you while dreaming up crazy experiments. The only limit of what you can do now is your imagination.

@mason_lab @JIWOONPARK_ @ACEpigenetics @cnag_eu @StJudeResearch @GESTALT_sp @ioavlachos
@brukerspatial @10xGenomics @vizgen_inc @AkoyaBio @ElemBio

[João Costa @jonnymates]

@JM_Sobral @CathyFPaulino Só se afogam em água limpa e pura, na merda navegam como ninguém. No leite estragado também rolam bem...

[João Costa @jonnymates]

@LoboLuaCheia @FranciscoMarkes Estava tudo excelente, nunca houve nada a criticar no passado, este homem sempre calado. O ordenado cerrava-lhe a voz

[João Costa @jonnymates]

@Sparky79 @RuiPinto_FL Tu parece que queres saber e te incomoda senão tavas caladinha

[João Costa @jonnymates]

@monta_bull @RuiPinto_FL Tás anémico? Com tuza parece que deixa de te chegar sangue ao cérebro. Vê isso

[João Costa @jonnymates]

@CurtaVermelha @RuiPinto_FL Coluna vermelha é sinal de inflamação. É de tanto dobrar. 10mg de honestidade e ficas bom

[João Costa @jonnymates]

@JM_Sobral 🤣

[João Costa @jonnymates]

The hard way... still waiting! 🤣

[Heng Li @lh3lh3]

Published in 2009, bwa-aln is one of the oldest short-read aligners. It remains a popular choice for aligning ancient DNA reads. The following blog post explains the algorithmic reasons. https://t.co/dhFrUtO0iE

[João Costa @jonnymates]

8 years ago today, Nanopore (and fish and chips) at the beach. Resulted in 2 (not that long) reads :).
@IGCiencia @IGCGenomics

[João Costa @jonnymates]

https://t.co/ry6QD4Y6Qx
8 years ago, Nanopore sequencing at the beach. 2 (not so long) reads was the result :)

[João Costa @jonnymates]

@gaby_gmartins "...hit me on the head with a hammer and everything that I’d worked for all my life was put into question...". Know the feeling, some people can take advantage of your work without blinking. Don't give up Sis!

[Patrick Hsu @pdhsu]

What if we could universally recombine, insert, delete, or invert any two pieces of DNA?

In back-to-back @Nature papers, we report the discovery of bridge RNAs and 3 atomic structures of the first natural RNA-guided recombinase - a new mechanism for programmable genome design

[Mile Sikic @msikic]

HERRO - imagine what you can do with accurate ultra-long nanopore reads https://t.co/K4CjHq3rYf w/ @domstanojevic @DehuiLin @PaolaFlorezdeS and great help from @sergeynurk #nanopore. See you at London Calling 2024!!

[João Costa @jonnymates]

One more expert in training! I'll hear from them all in the end... @IGCGenomics

[Nature Reviews Genetics @NatureRevGenet]

Rapid pathogen surveillance: field-ready sequencing solutions https://t.co/PqtOaH7lCU #JournalClub by @kirstynbrunker @UofGlasgow