Dataset Information

A generalizable Cas9/sgRNA prediction model using machine transfer learning with small high-quality datasets.

ABSTRACT: The CRISPR/Cas9 nuclease from Streptococcus pyogenes (SpCas9) can be used with single guide RNAs (sgRNAs) as a sequence-specific antimicrobial agent and as a genome-engineering tool. However, current bacterial sgRNA activity models struggle with accurate predictions and do not generalize well, possibly because the underlying datasets used to train the models do not accurately measure SpCas9/sgRNA activity and cannot distinguish on-target cleavage from toxicity. Here, we solve this problem by using a two-plasmid positive selection system to generate high-quality data that more accurately reports on SpCas9/sgRNA cleavage and that separates activity from toxicity. We develop a machine learning architecture (crisprHAL) that can be trained on existing datasets, that shows marked improvements in sgRNA activity prediction accuracy when transfer learning is used with small amounts of high-quality data, and that can generalize predictions to different bacteria. The crisprHAL model recapitulates known SpCas9/sgRNA-target DNA interactions and provides a pathway to a generalizable sgRNA bacterial activity prediction tool that will enable accurate antimicrobial and genome engineering applications.

SUBMITTER: Ham DT

PROVIDER: S-EPMC10485023 | biostudies-literature | 2023 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A generalizable Cas9/sgRNA prediction model using machine transfer learning with small high-quality datasets.

Ham Dalton T DT Browne Tyler S TS Banglorewala Pooja N PN Wilson Tyler L TL Michael Richard K RK Gloor Gregory B GB Edgell David R DR

Nature communications 20230907 1

The CRISPR/Cas9 nuclease from Streptococcus pyogenes (SpCas9) can be used with single guide RNAs (sgRNAs) as a sequence-specific antimicrobial agent and as a genome-engineering tool. However, current bacterial sgRNA activity models struggle with accurate predictions and do not generalize well, possibly because the underlying datasets used to train the models do not accurately measure SpCas9/sgRNA activity and cannot distinguish on-target cleavage from toxicity. Here, we solve this problem by usi ...[more]

PMID: 37679324

Similar Datasets

Project description:BackgroundDepressive symptoms are rising in the general population, but their associated factors are unclear. Although the link between sleep disturbances and depressive symptoms severity (DSS) is reported, the predictive role of sleep on DSS and the impact of anxiety and the brain on their relationship remained obscure.MethodsUsing three population-based datasets (N = 1813), we trained the machine learning models in the primary dataset (N = 1101) to assess the predictive role of sleep quality, anxiety problems, and brain structural (and functional) measurements on DSS, then we tested our models' performance in two independent datasets (N = 378, N = 334) to test the generalizability of our findings. Furthermore, we applied our model to a smaller longitudinal subsample (N = 66). In addition, we performed a mediation analysis to identify the role of anxiety and brain measurements on the sleep quality and DSS association.FindingsSleep quality could predict individual DSS (r = 0.43, R2 = 0.18, rMSE = 2.73), and adding anxiety, contrary to brain measurements, strengthened its prediction performance (r = 0.67, R2 = 0.45, rMSE = 2.25). Importantly, out-of-cohort validations in other cross-sectional datasets and a longitudinal subsample provided robust similar results. Furthermore, anxiety scores, contrary to brain measurements, mediated the association between sleep quality and DSS.InterpretationPoor sleep quality could predict DSS at the individual subject level across three datasets. Anxiety scores not only increased the predictive model's performance but also mediated the link between sleep quality and DSS.FundingThe study is supported by Helmholtz Imaging Platform grant (NimRLS, ZTI-PF-4-010), the Deutsche Forschungsgemeinschaft (DFG, GE 2835/2-1, GE 2835/4-1), the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation)-Project-ID 431549029-SFB 1451, the programme "Profilbildung 2020" (grant no. PROFILNRW-2020-107-A), an initiative of the Ministry of Culture and Science of the State of Northrhine Westphalia.

Project description:Over the last three decades, the appeal for monoclonal antibodies (mAbs) as therapeutics has been steadily increasing as evident with FDA's recent landmark approval of the 100th mAb. Unlike mAbs that bind to single targets, multispecific biologics (msAbs) have garnered particular interest owing to the advantage of engaging distinct targets. One important modular component of msAbs is the single-chain variable fragment (scFv). Despite the exquisite specificity and affinity of these scFv modules, their relatively poor thermostability often hampers their development as a potential therapeutic drug. In recent years, engineering antibody sequences to enhance their stability by mutations has gained considerable momentum. As experimental methods for antibody engineering are time-intensive, laborious and expensive, computational methods serve as a fast and inexpensive alternative to conventional routes. In this work, we show two machine learning approaches - one with pre-trained language models (PTLM) capturing functional effects of sequence variation, and second, a supervised convolutional neural network (CNN) trained with Rosetta energetic features - to better classify thermostable scFv variants from sequence. Both of these models are trained over temperature-specific data (TS50 measurements) derived from multiple libraries of scFv sequences. On out-of-distribution (refers to the fact that the out-of-distribution sequnes are blind to the algorithm) sequences, we show that a sufficiently simple CNN model performs better than general pre-trained language models trained on diverse protein sequences (average Spearman correlation coefficient, ρ, of 0.4 as opposed to 0.15). On the other hand, an antibody-specific language model performs comparatively better than the CNN model on the same task (ρ= 0.52). Further, we demonstrate that for an independent mAb with available thermal melting temperatures for 20 experimentally characterized thermostable mutations, these models trained on TS50 data could identify 18 residue positions and 5 identical amino-acid mutations showing remarkable generalizability. Our results suggest that such models can be broadly applicable for improving the biological characteristics of antibodies. Further, transferring such models for alternative physicochemical properties of scFvs can have potential applications in optimizing large-scale production and delivery of mAbs or bsAbs.

Dataset Information

A generalizable Cas9/sgRNA prediction model using machine transfer learning with small high-quality datasets.

Publications

A generalizable Cas9/sgRNA prediction model using machine transfer learning with small high-quality datasets.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets