Genie TechBio Inc.’s Post

I've written up a checklist to ask before starting any data or bioinformatics project. Four questions I've learned to ask, hopefully they'll help you! https://lnkd.in/gb55UWtP

Happy to share that my FIRST paper as first author has been published by Scientific Reports. I’m truly grateful to my supervisors Zohar Yakhini and Leon Anavy for their undivided attention, guidance, and continued support every step of the way, to Michael Rosenberg for our collaboration, and to the Yakhini Research Group for our work together. Thank you all for making this happen. The paper is called, Efficient DNA-based data storage using shortmer combinatorial encoding, and you can check it out, here: https://lnkd.in/dygKrEs5 #NaturePortfolio #ScientificReports #DNADataStorage #DataStorage #CombinatorialEncoding

Lesson 4: Be prepared to learn the domain! Having conducted data science research in the biomedical field, I’ve learned that data science is an applied discipline. On its own, it's just a collection of mathematical concepts, but its true value comes out when applied to a specific domain to uncover insights. For instance, in my research, I needed to understand various processes like translation, transcription, RNA structure, and protein synthesis to truly understand the work. While I didn’t have to become an expert in these areas, having this foundational knowledge made me feel more confident in what I was doing and gave me a better intuition. Understanding the context behind what you're doing builds confidence and keeps you motivated. #LearningDataScience #MachineLearning

🔬 SnapATAC2: A New Tool for Single-Cell Omics Analysis 🔬 A recent study in Nature Methods (https://buff.ly/3vUCEVd) unveils SnapATAC2, a new Python tool revolutionizing single-cell epigenomics analysis. 🌐 Key Features: 🔵 Advanced Algorithm: SnapATAC2 employs a matrix-free spectral embedding algorithm, enhancing the analysis of cellular heterogeneity and scalability for large datasets. 🔵 Efficient Dimensionality Reduction: It projects high-dimensional single-cell omics data into low-dimensional space, preserving cell relationships, crucial for understanding complex biological systems. 🔵 Innovative Use of Lanczos Algorithm: This approach bypasses the need for constructing a graph Laplacian matrix, significantly reducing storage space requirements. 🔵 Wide array of single-cell omics data, including ATAC-seq, RNA-seq, Hi-C, and methylation. 🔵 Comprehensive Functionalities: The tool encompasses preprocessing, embedding/clustering, functional enrichment analysis, and multimodal omics analysis. 👩💻 Github page: https://buff.ly/3OjgaDG 🔍 SnapATAC2 stands out for its ability to handle the computational challenges in single-cell omics data analysis, making it a valuable asset for biologists, bioinformaticians, and pharma leaders. 📢 Join the Conversation 📢 Share your thoughts, methods, and tools in the comments 👇 💬  #SingleCellAnalysis #Epigenomics #Bioinformatics #SnapATAC2 #PythonTools #Genomics #DataScience

Bioinformatics Journey: 004 In a data science project, the size of the dataset matters a lot. For this drug discovery project of mine, I mistakenly used small dataset (it's also the only one available in the ChEMBL dataset with Homo sapiens as the target organism). This made the model building difficult, and since it's a biological data, I can't just treat it like other data and perform data augmentation. Having domain knowledge is required for it. That's why, I've changed the dataset and looked for another target protein: norepinephrine transporter (NET). And... it worked! The R-squared values are finally not on the negative side! Hehe Overall, this project taught me a lot, both from the biological and data side. Lipinski's rule of five is an interesting concept I've learned from this too! You can check the repository here: https://lnkd.in/gxzUrjjR Included are the biological concepts I've learned along the way. I had fun working on this project, looking forward to more in the future :> #beyondthevinculum #bioinformatics #datascience #drugdiscovery

I'm really glad to announce that my first first-author paper has been published in Bioinformatics, "AutoPeptideML: a study on how to build more trustworthy peptide bioactivity predictors". The main idea behind the paper is to build a tool to automate machine learning modelling for peptide bioactivity prediction, so that any experimental scientist can easily build their own custom models and use them to accelerate their research, without a single line of code. The tool is completely open source and a webserver is available for anyone to use. This paper could not have been possible without the help of my supervisors Denis Shields, .Hoang .Thanh Lam, and Vanessa Lopez; and my colleagues Clement Agoni, PhD and Rodrigo Cossio-Pérez. If you want to know more or use the tool, here are some useful links: - 📖 Paper: https://lnkd.in/dFvQZCEi - 🛠️ Github Repository: https://lnkd.in/dCkx9N6V - 🚀 Webserver: https://lnkd.in/dqdBiyzi

hashtag #Starts_in_4_days - Command line Bioinformatics for Molecular Biology Register here - https://lnkd.in/daXyJRvc Check reviews here - https://lnkd.in/de8_bKQa In this basic course, I introduce you to command line bioinformatics for common tasks in molecular Biology - e.g. Sequence Retrieval, Analysis, Primer Designing, Homology Search etc. I introduce you to general features and utility of EMBOSS - The European Molecular Biology Open Software Suite - by demonstrating the utility of some of the common tools in EMBOSS - including Seqret, Sixpack, Restrict etc. Besides, we also demonstrate the utility of Awk, Sed and grep grep - stands for global regular expression search and print)  is a command-line utility for searching plaintext datasets for lines that match a regular expression. Awk - is a general-purpose scripting language designed for advanced text processing and parsing large files SED - is a powerful text stream editor. Can do insertion, deletion, search and replace(substitution). grep, awk and sed together form the proverbial swiss knife of data analytics.

Bioinformatics: Data Analysis A Venn diagram is a widely used diagram style that shows the logical relation between sets, popularized by John Venn in the 1880s. The diagrams are used to teach elementary set theory, and to illustrate simple set relationships in probability, logic, statistics, linguistics and computer science. A Venn diagram uses simple closed curves drawn on a plane to represent sets. Very often, these curves are circles or ellipses. Problem: In a family of enzymes, 15 participate in a type I reaction, 12 in a type II reaction, 5 in both and 8 have no activity in a type I or II. How many total enzymes are in this family? A\) 30 B\) 27 C\) 40 D\) 22 E\) 35 Youtube video: https://lnkd.in/g5K7Fb8D \#nikolays_genetics_lessons

Survival of the Fittest Programs: How Machines Evolve to Solve Problems with Genetic Programming via #TowardsAI → https://bit.ly/3YiRSPx

Yesterday, the Computational Population Genetics group discussed the most recent Deep Learning Architectures developed for the genetics field. It was very exciting to learn about new methodologies and how they have been applied in different contexts with a large amount of data. This is very important for future studies, where NGS technologies are yielding more and more information, and very few scientists (half bioinformaticians and half machine learning engineers) are applying these methods. In conclusion, we are planning to create a mini-workshop with projects for each member. Wait for the results in GitHub! https://lnkd.in/eRd3gr44