LatchBio

Because many classes of analysis problems in biology involve searching, manipulating or otherwise moving around large amounts of file data, progress in systems programming will have a large impact on therapeutics development and basic biology research alike. But while the pace of data generation from experimental techniques is increasing, there are actually few assays where the computational step is the bottleneck in the end to end research workflow. In industry, high quality open source tools and widespread access to cloud infrastructure are often sufficient to distill GBs of raw data into small tables of values and plots with clear biological interpretation. If we consider bulk RNA-seq and genetic sequencing as two representative examples, both have well-maintained aligners with strong consensus amongst computational biologists. So even when the data size becomes quite large, as deep WGS can yield raw files 100s of GBs in size, the use of computers is a small component of a workflow dominated by the wet lab Spatial single cell epigenetics is one such experimental technique where the existing software tools simply break and computational steps are rate limiting towards interpretable results. This assay provides the biomolecular state of tissue with geometric structure, building beautiful pictures of cell state progression, gene programs and mechanisms that elucidate disease pathogenesis and basic biology. Unfortunately, the multi-modal nature of spatial epigenetics and the volume of data generated from modern kits is actually breaking the fragmented ecosystem of tools that has emerged to process it. Most analysis workflows of this data type lean heavily on core objects and operations defined in the Seurat or Scanpy packages, popular frameworks for the exploratory analysis of single cell data. These codebases were born in academic biology labs and started as small Python and R projects for the early days of single cell biology. Spatial assays routinely generate >1M cells worth of sequencing data and this amount is doubling every few years as assay developers are shrinking the feature size on tissue slides with parallels to semiconductor manufacturing decades prior. Working with this volume of single cell data is very difficult and reveals the seams in the existing academic codebases. We are sitting down with AtlasXomics Inc., a talented group of scientists and engineers that are building these kits. If you are a strong programmer and want to make a contribution to the pace of experimental biology in practice, this is a good place to look. If you are building a biotech platform, or working in the ivory tower, want to understand what state-of-the-art spatial workflows look like and the rich biology they uncover, please tune in. https://lnkd.in/gEVjCmG2

Announcing full support for Fluent Biosciences single-cell kits on LatchBio 🧬 Over the past 6 months, we’ve partnered with the Fluent team to empower their customers to self-serve data analysis for all single-cell kits, proving out: 2k cells 10k cells 20k cells 100k cells 1000k cells In our largest experiment, a Morris Lab PhD candidate at Washington University, St. Louis ran 1 million cell sequencing for cellular reprogramming. This involved processing 7 billion reads in under 72 hours on latch 🤯 (more on this soon...) Scientists are seeing a decrease in time to insight and processing costs of up to 80-95% across all kit sizes. And wet lab biologists are actually able to use it. :) Today, anyone can run PIP-Seq analysis in a way that is easy to use and requires no coding. See the link and case studies in the comments below 👇 We’re excited to continue supporting Fluent as they pioneer single-cell biology at Illimuna. Thank you for continuous support from Fluent BioSciences Inc., Kristina Fontanez, Sadie VanHorn, Yigal Agam, Aaron May-Zhang, PhD, Robert Meltzer, Jessie Matakis, Rahul Desai, Kenny Workman, Emma Krivoshein, and Aidan Abdulali. All of your involvement, development, and feedback made this possible. We're committed to continue driving down costs and making single cell biology - and bioinformatics - scalable and accessible to everyone.

AtlasXomics Inc. is creating beautiful maps at cellular resolution to reveal the epigenetic underpinnings of disease development. Creating these maps generates 100s of TBs of data and requires a vast computational infrastructure. Tune in to our webinar on Thursday, August 1st to hear from Colin Ng, VP of AtlasXomics, about how they partner with LatchBio to overcome this data challenge and deliver quicker insights to their customers. Register here to join: https://lnkd.in/gMPmWATC

We are thrilled to host a webinar with one of our partners, AtlasXomics Inc.! As a pioneer in spatial biology technologies, AtlasXomics offers the first and only comprehensive spatial epigenome solution at a cellular resolution. Their platform is now used by over 50 labs to uncover the epigenetic drivers behind key disease areas. Join us on Thursday, August 1st to hear from Colin Ng, VP at AtlasXOmics, as he discusses the development of their breakthrough platform and how they’ve utilized LatchBio to unlock key insights for their customers. We hope to see you there! Register here to tune in: https://lnkd.in/gMPmWATC

Congratulations to our partner Fluent BioSciences Inc. on their acquisition by Illumina. The future of single-cell technology is bright.

Latch is now SOC 2 Type II compliant! Our team, software, and procedures demonstrated best-in-class security and data privacy practices during the three-month independent audit, including penetration tests. Cloud security is rightfully a major concern for biotechs modernizing data infrastructure. Visit our Trust Center for thorough overviews of our security and operational safeguards at https://trust.latch.bio

Congrats to EvolutionaryScale for the release of ESM3, an LLM that uses sequence, structure and function to generate new proteins. The model is now on Latch for non-commercial use and gives the academic community access to their cutting-edge industrial research with a few clicks.

Congrats to EvolutionaryScale for the release of ESM3, an LLM that uses sequence, structure and function to generate new proteins. The model is now on Latch for non-commercial use and gives the academic community access to their cutting-edge industrial research with a few clicks.

The CSO should drive the deployment and regular use of data platforms within a biotech organization rather than a distinct engineering lead. This is both because the requirements for the platform should be set by the biological goals of the company and adopting such a platform in practice requires strong scientific leadership to change the daily behavior of researchers. The current standard is to delegate this responsibility to a computational lead who is further removed from the scientific goals of the organization by training and detached from the daily behavior and needs of the end users. This has led to the widespread deployment of bloated, buggy, and often disjointed platforms that are never truly adopted by research teams and take many years and resources to stand up. Thus the research productivity gained from a single collaborative environment for access and exploration of all experimental data over time is simply unrealized for many biotech organizations. These general claims are made from observing the internals of hundreds of biotech companies. CSO is used interchangeably with scientific leadership throughout this essay and describes the person or team that plans drug programs.

Yesterday we presented the first map of the Omics Solution Provider landscape. Many people asked me for the extra insights so I'll just share them here. Starting with geography! The most represented cities in the dataset were relatively expected, but with a surprising tie for second: 1. San Diego 2. Cambridge, Boston, San Jose 3. South San Francisco For the most represented countries there was a 10x difference between first and second: 1. United States 2. United Kingdom 3. Switzerland And of course the old age discussion: What if we group cities into their metropolitan areas? 1. Bay Area (blessed be thy bay) 2. Boston Area 3. San Diego Area - This dataset is a work in progress, we expect to rapidly update it as we learn more and new companies get started. DM me (or comment) and i'm happy to share it with you.