Predicting willingness to donate blood based on machine learning: two blood donor recruitments during COVID-19 outbreaks.

Machine learning methods are a novel way to predict and rank donors' willingness to donate blood and to achieve precision recruitment, which can improve the recruitment efficiency and meet the challenge of blood shortage. We collected information about experienced blood donors via short message service (SMS) recruitment and developed 7 machine learning-based recruitment models using PyCharm-Python Environment and 13 features which were described as a method for ranking and predicting donors' intentions to donate blood with a floating number between 0 and 1. Performance of the prediction models was assessed by the Area under the receiver operating characteristic curve (AUC), accuracy, precision, recall, and F1 score in the full dataset, and by the accuracy in the four sub-datasets. The developed models were applied to prospective validations of recruiting experienced blood donors during two COVID-19 pandemics, while the routine method was used as a control. Overall, a total of 95,476 recruitments via SMS and their donation results were enrolled in our modelling study. The strongest predictor features for the donation of experienced donors were blood donation interval, age, and donation frequency. Among the seven baseline models, the eXtreme Gradient Boosting (XGBoost) and Support vector machine models (SVM) achieved the best performance: mean (95%CI) with the highest AUC: 0.809 (0.806-0.811), accuracy: 0.815 (0.812-0.818), precision: 0.840 (0.835-0.845), and F1 score of XGBoost: 0.843 (0.840-0.845) and recall of SVM: 0.991 (0.988-0.994). The hit rate of the XGBoost model alone and the combined XGBoost and SVM models were 1.25 and 1.80 times higher than that of the conventional method as a control in 2 recruitments respectively, and the hit rate of the high willingness to donate group was 1.96 times higher than that of the low willingness to donate group. Our results suggested that the machine learning models could predict and determine the experienced donors with a strong willingness to donate blood by a ranking score based on personalized donation data and demographical details, significantly improve the recruitment rate of blood donors and help blood agencies to maintain the blood supply in emergencies.

Electrochemical Detection of a Few Copies of Unamplified SARS-CoV-2 Nucleic Acids by a Self-Actuated Molecular System.

The existing electrochemical biosensors lack controllable and intelligent merit to modulate the sensing process upon external stimulus, leading to challenges in analyzing a few copies of biomarkers in unamplified samples. Here, we present a self-actuated molecular-electrochemical system that consists of a tentacle and a trunk modification on a graphene microelectrode. The tentacle that contains a probe and an electrochemical label keeps an upright orientation, which increases recognition efficiency while decreasing the pseudosignal. Once the nucleic acids are recognized, the tentacles nearby along with the labels are spontaneously actuated downward, generating electrochemical responses under square wave voltammetry. Thus, it detects unamplified SARS-CoV-2 RNAs within 1 min down to 4 copies in 80 µL, 2-6 orders of magnitude lower than those of other electrochemical assays. Double-blind testing and 10-in-1 pooled testing of nasopharyngeal samples yield high overall agreement with reverse transcription-polymerase chain reaction results. We fabricate a portable prototype based on this system, showing great potential for future applications.

Triple-Probe DNA Framework-Based Transistor for SARS-CoV-2 10-in-1 Pooled Testing.

Accurate and population-scale screening technology is crucial in the control and prevention of COVID-19, such as pooled testing with high overall testing efficiency. Nevertheless, pooled testing faces challenges in sensitivity and specificity due to diluted targets and increased contaminations. Here, we develop a graphene field-effect transistor sensor modified with triple-probe tetrahedral DNA framework (TDF) dimers for 10-in-1 pooled testing of SARS-CoV-2 RNA. The synergy effect of triple probes as well as the special nanostructure achieve a higher binding affinity, faster response, and better specificity. The detectable concentration reaches 0.025-0.05 copy µL-1 in unamplified samples, lower than that of the reverse transcript-polymerase chain reaction. Without a requirement of nucleic-acid amplification, the sensors identify all of the 14 positive cases in 30 nasopharyngeal swabs within an average diagnosis time of 74 s. Unamplified 10-in-1 pooled testing enabled by the triple-probe TDF dimer sensor has great potential in the screening of COVID-19 and other epidemic diseases.

Rapid and ultrasensitive electromechanical detection of ions, biomolecules and SARS-CoV-2 RNA in unamplified samples.

The detection of samples at ultralow concentrations (one to ten copies in 100 µl) in biofluids is hampered by the orders-of-magnitude higher amounts of 'background' biomolecules. Here we report a molecular system, immobilized on a liquid-gated graphene field-effect transistor and consisting of an aptamer probe bound to a flexible single-stranded DNA cantilever linked to a self-assembled stiff tetrahedral double-stranded DNA structure, for the rapid and ultrasensitive electromechanical detection (down to one to two copies in 100 µl) of unamplified nucleic acids in biofluids, and also of ions, small molecules and proteins, as we show for Hg2+, adenosine 5'-triphosphate and thrombin. We implemented an electromechanical biosensor for the detection of SARS-CoV-2 into an integrated and portable prototype device, and show that it detected SARS-CoV-2 RNA in less than four minutes in all nasopharyngeal samples from 33 patients with COVID-19 (with cycle threshold values of 24.9-41.3) and in none of the 54 COVID-19-negative controls, without the need for RNA extraction or nucleic acid amplification.

Ultraprecise Antigen 10-in-1 Pool Testing by Multiantibodies Transistor Assay.

Effective screening of infectious diseases requires a fast, cheap, and population-scale testing. Antigen pool testing can increase the test rate and shorten the screening time, thus being a valuable approach for epidemic prevention and control. However, the overall percent agreement (OPA) with polymerase chain reaction (PCR) is one-half to three-quarters, hampering it from being a comprehensive method, especially pool testing, beyond the gold-standard PCR. Here, a multiantibodies transistor assay is developed for sensitive and highly precise antigen pool testing. The multiantibodies capture SARS-CoV-2 spike S1 proteins with different configurations, resulting in an antigen-binding affinity down to 0.34 fM. The limit of detection reaches 3.5 × 10-17 g mL-1SARS-CoV-2 spike S1 protein in artificial saliva, 4-5 orders of magnitude lower than existing transistor sensors. The testing of 60 nasopharyngeal swabs exhibits ∼100% OPA with PCR within an average diagnoses time of 38.9 s. Owing to its highly precise feature, a portable integrated platform is fabricated, which achieves 10-in-1 pooled screening for high testing throughput. This work solves the long-standing problem of antigen pool testing, enabling it to be a valuable tool in precise diagnoses and population-wide screening of COVID-19 or other epidemics in the future.

Rapid SARS-CoV-2 Nucleic Acid Testing and Pooled Assay by Tetrahedral DNA Nanostructure Transistor.

Direct SARS-CoV-2 nucleic acid testing with fast speed and high frequency is crucial for controlling the COVID-19 pandemic. Here, direct testing of SARS-CoV-2 nucleic acid is realized by field-effect transistors (FETs) with an electro-enrichable liquid gate (LG) anchored by tetrahedral DNA nanostructures (TDNs). The applied gate bias electrostatically preconcentrates nucleic acids, while the liquid gate with TDNs provides efficient analyte recognition and signal transduction. The average diagnosis time is ∼80 s, and the limit of detection approaches 1-2 copies in 100 µL of clinical samples without nucleic acid extraction and amplification. As such, TDN-LG FETs solve the dilemma of COVID-19 testing on mass scale that diagnosis accuracy and speed undergo trade-off. In addition, TDN-LG FETs achieve unamplified 10-in-1 pooled nucleic acid testing for the first time, and the results are consistent with PCR. Thus, this technology promises on-site and wide population COVID-19 screening and ensures safe world-reopening.

Direct SARS-CoV-2 Nucleic Acid Detection by Y-Shaped DNA Dual-Probe Transistor Assay.

Rapid screening of infected individuals from a large population is an effective means in epidemiology, especially to contain outbreaks such as COVID-19. The gold standard assays for COVID-19 diagnostics are mainly based on the reverse transcription polymerase chain reaction, which mismatches the requirements for wide-population screening due to time-consuming nucleic acid extraction and amplification procedures. Here, we report a direct nucleic acid assay by using a graphene field-effect transistor (g-FET) with Y-shaped DNA dual probes (Y-dual probes). The assay relies on Y-dual probes modified on g-FET simultaneously targeting ORF1ab and N genes of SARS-CoV-2 nucleic acid, enabling high a recognition ratio and a limit of detection (0.03 copy µL-1) 1-2 orders of magnitude lower than existing nucleic acid assays. The assay realizes the fastest nucleic acid testing (∼1 min) and achieves direct 5-in-1 pooled testing for the first time. Owing to its rapid, ultrasensitive, easily operated features as well as capability in pooled testing, it holds great promise as a comprehensive tool for population-wide screening of COVID-19 and other epidemics.

Ultrasensitive Detection of SARS-CoV-2 Antibody by Graphene Field-Effect Transistors.

The fast spread of SARS-CoV-2 has severely threatened the public health. Establishing a sensitive method for SARS-CoV-2 detection is of great significance to contain the worldwide pandemic. Here, we develop a graphene field-effect transistor (g-FET) biosensor and realize ultrasensitive SARS-CoV-2 antibody detection with a limit of detection (LoD) down to 10-18 M (equivalent to 10-16 g mL-1) level. The g-FETs are modified with spike S1 proteins, and the SARS-CoV-2 antibody biorecognition events occur in the vicinity of the graphene surface, yielding an LoD of ∼150 antibodies in 100 µL full serum, which is the lowest LoD value of antibody detection. The diagnoses time is down to 2 min for detecting clinical serum samples. As such, the g-FETs leverage rapid and precise SARS-CoV-2 screening and also hold great promise in prevention and control of other epidemic outbreaks in the future.